U.S. patent application number 10/280376 was filed with the patent office on 2003-12-11 for novel microarrays and methods of use thereof.
This patent application is currently assigned to The Trustees of Columbia University in the City of New York. Invention is credited to Wang, Denong.
Application Number | 20030228637 10/280376 |
Document ID | / |
Family ID | 23083717 |
Filed Date | 2003-12-11 |
United States Patent
Application |
20030228637 |
Kind Code |
A1 |
Wang, Denong |
December 11, 2003 |
Novel microarrays and methods of use thereof
Abstract
This invention provides novel nitrocellulose-based or
Hydrogel-based microarrays and methods of making and using them (1)
to detect the presence of one or more agents in a sample, (2) to
determine the amount of one or more agents in a sample, (3) to
determine whether a subject is afflicted with a disorder, and (4)
to determine whether an agent known to specifically bind to a first
compound also specifically binds to a second compound. This
invention also provides kits which comprise the instant
microarrays. This invention further provides antibodies capable of
specifically binding to a glycomer present both on the surface of a
mammalian macrophage or intestinal epithelial cell, and on a
bacterial cell. Finally, this invention provides diagnostic methods
using the instant antibodies.
Inventors: |
Wang, Denong; (Middletown,
NY) |
Correspondence
Address: |
John P. White, Esq.
Cooper & Dunham, LLP
23rd Floor
1185 Avenue of the Americas
New York
NY
10036
US
|
Assignee: |
The Trustees of Columbia University
in the City of New York
|
Family ID: |
23083717 |
Appl. No.: |
10/280376 |
Filed: |
October 24, 2002 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10280376 |
Oct 24, 2002 |
|
|
|
PCT/US02/11612 |
Apr 10, 2002 |
|
|
|
60282926 |
Apr 10, 2001 |
|
|
|
Current U.S.
Class: |
435/7.9 ;
435/287.2 |
Current CPC
Class: |
B01J 19/0046 20130101;
B01J 2219/00626 20130101; G01N 2400/00 20130101; B01J 2219/00727
20130101; B01J 2219/00527 20130101; B01L 3/5085 20130101; B01J
2219/00725 20130101; G01N 33/548 20130101; B01J 2219/005 20130101;
C40B 40/10 20130101; B01J 2219/00621 20130101; B01J 2219/00605
20130101; B01J 2219/00617 20130101; B01J 2219/00641 20130101; B01J
2219/00612 20130101; B01J 2219/00644 20130101; G01N 33/5308
20130101; B01J 2219/0074 20130101; B01J 2219/00637 20130101 |
Class at
Publication: |
435/7.9 ;
435/287.2 |
International
Class: |
G01N 033/53; G01N
033/542; C12M 001/34 |
Claims
What is claimed is:
1. A microarray comprising a nitrocellulose or Hydrogel support
having affixed to its surface at discrete loci a plurality of
compounds, wherein (a) at at least one discrete locus is affixed a
compound selected from the group consisting of a glycomer, an
insoluble protein, a lectin and an antibody, and (b) the
composition of compounds at each discrete locus differs from the
composition of compounds at at least one other discrete locus.
2. The microarray of claim 1, wherein the nitrocellulose or
Hydrogel support is selected from the group consisting of a chip, a
slide, a filter, and a plate.
3. A microarray comprising a plurality of nitrocellulose or
Hydrogel or Hydrogel supports, each support having one or a
plurality of compounds affixed to its surface at a single discrete
locus or a plurality of compounds affixed to its surface at
discrete loci, wherein (a) at at least one discrete locus is
affixed a compound selected from the group consisting of a
glycomer, an insoluble protein, a lectin and an antibody, and (b)
the composition of compounds at each discrete locus differs from
the composition of compounds at at least one other discrete
locus.
4. The microarray of claim 3, wherein the nitrocellulose or
Hydrogel or Hydrogel support is selected from the group consisting
of a chip, a slide, a filter, a plate, and a bead.
5. The microarray of claim 1 or 3, wherein the number of discrete
loci is at least 100.
6. The microarray of claim 1 or 3, wherein the number of discrete
loci is at least 1000.
7. The microarray of claim 1 or 3, wherein the number of discrete
loci is at least 10,000.
8. The microarray of claim 1 or 3, wherein a glycomer is affixed at
at least one locus.
9. The microarray of claim 1 or 3, wherein an insoluble protein is
affixed at at least one locus.
10. The microarray of claim 1 or 3, wherein a lectin is affixed at
at least one locus.
11. The microarray of claim 1 or 3, wherein an antibody is affixed
at at least one locus.
12. The microarray of claim 1 or 3, wherein at each locus is
affixed only one compound.
13. The microarray of claim 1 or 3, wherein at at least one locus
is affixed a plurality of compounds.
14. The microarray of claim 1 or 3, wherein the microarray has
affixed to its surface two or more compounds selected from the
group consisting of a glycomer, an insoluble protein, a lectin and
an antibody.
15. The microarray of claim 1 or 3, wherein the microarray has
further affixed to its surface a compound selected from the group
consisting of a soluble protein, a nucleic acid and a small
molecule.
16. An article comprising a nitrocellulose or Hydrogel support
having dextran affixed to its surface at discrete loci.
17. The article of claim 16, wherein the dextran is .alpha.(1,6)
dextran.
18. A microarray comprising the article of claim 16, wherein at
least one compound is affixed to the dextran at each discrete
locus, the composition of compounds at each discrete locus
differing from the composition of compounds at at least one other
discrete locus.
19. The microarray of claim 18, wherein the nitrocellulose or
Hydrogel support is selected from the group consisting of a chip, a
slide, a filter, and a plate.
20. An article comprising a plurality of nitrocellulose or Hydrogel
supports, each support having dextran affixed to its surface at one
or more discrete loci.
21. The article of claim 20, wherein the dextran is .alpha.(1,6)
dextran.
22. A microarray comprising the article of claim 20, wherein at
least one compound is affixed to the dextran at each discrete
locus, the composition of compounds at each discrete locus
differing from the composition of compounds at at least one other
discrete locus.
23. The microarray of claim 22, wherein the nitrocellulose or
Hydrogel support is selected from the group consisting of a chip, a
slide, a filter, a plate, and a bead.
24. The microarray of claim 18 or 22, wherein the number of
discrete loci is at least 100.
25. The microarray of claim 18 or 22, wherein the number of
discrete loci is at least 1000.
26. The microarray of claim 18 or 22, wherein the number of
discrete loci is at least 10,000.
27. The microarray of claim 18 or 22, wherein a glycomer is affixed
to the dextran at at least one locus.
28. The microarray of claim 18 or 22, wherein an insoluble protein
is affixed to the dextran at at least one locus.
29. The microarray of claim 18 or 22, wherein a lectin is affixed
to the dextran at at least one locus.
30. The microarray of claim 18 or 22, wherein an antibody is
affixed to the dextran at at least one locus.
31. The microarray of claim 18 or 22, wherein at each locus is
affixed only one compound.
32. The microarray of claim 18 or 22, wherein at at least one locus
is affixed a plurality of compounds.
33. The microarray of claim 18 or 22, wherein the microarray has
affixed to the dextran two or more compounds selected from the
group consisting of a glycomer, an insoluble protein, a lectin and
an antibody.
34. The microarray of claim 18 or 22, wherein the microarray has
affixed to its surface a compound selected from the group
consisting of a soluble protein, a nucleic acid and a small
molecule.
35. A method of detecting in a sample the presence of one or more
agents which specifically bind to one or more known glycomers,
which method comprises: (a) contacting the sample with the
microarray of claim 1 or 3, wherein each known glycomer is affixed
at at least one discrete locus and wherein the contacting is
performed under conditions which would permit an agent, if present
in the sample, to specifically bind to its corresponding glycomer
in the microarray; and (b) determining whether any known glycomer
in the microarray has an agent specifically bound thereto, thereby
detecting the presence of the one or more agents in the sample.
36. The method of claim 35, wherein the agent is an antibody which
correlates with an inflammatory disease.
37. The method of claim 35, wherein the agent is an antibody which
correlates with an infection.
38. The method of claim 35, wherein the agent is an antibody which
correlates with the presence of a tumor.
39. The method of claim 35, wherein the method comprises detecting
the presence of a plurality of agents in the sample, each of which
binds to a plurality of glycomers.
40. The method of claim 35, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to one glycomer.
41. A method of detecting in a sample the presence of one or more
agents which specifically bind to one or more known insoluble
proteins, which method comprises: (a) contacting the sample with
the microarray of claim 1 or 3, wherein each known insoluble
protein is affixed at at least one discrete locus and wherein the
contacting is performed under conditions which would permit an
agent, if present in the sample, to specifically bind to its
corresponding insoluble protein in the microarray; and (b)
determining whether any known insoluble protein in the microarray
has an agent specifically bound thereto, thereby detecting the
presence of the one or more agents in the sample.
42. The method of claim 41, wherein the agent is an antibody which
correlates with a disease.
43. The method of claim 41, wherein the agent is an antibody which
correlates with an infection.
44. The method of claim 41, wherein the agent is an antibody which
correlates with the presence of a tumor.
45. The method of claim 41, wherein the method comprises detecting
the presence of a plurality of agents in the sample, each of which
binds to a plurality of insoluble proteins.
46. The method of claim 41, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to one insoluble protein.
47. A method of detecting in a sample the presence of one or more
agents which specifically bind to one or more known antibodies or
lectins, which method comprises: (a) contacting the sample with the
microarray of claim 1 or 3, wherein each known antibody or lectin
is affixed at at least one discrete locus and wherein the
contacting is performed under conditions which would permit an
agent, if present in the sample, to specifically bind to its
corresponding antibody or lectin in the microarray; and (b)
determining whether any known antibody or lectin in the microarray
has an agent specifically bound thereto, thereby detecting the
presence of the one or more agents in the sample.
48. The method of claim 47, wherein the agent is an antibody which
correlates with a disease.
49. The method of claim 47, wherein the agent is an antibody which
corresponds to an infection.
50. The method of claim 47, wherein the agent is an antibody which
correlates with the presence of a tumor.
51. The method of claim 47, wherein the method comprises detecting
the presence of a plurality of agents in the sample, each of which
binds to a plurality of lectins or antibodies.
52. The method of claim 47, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to one lectin or antibody.
53. A method of determining the amount of one or more agents in a
sample, each of which specifically binds to one or more known
glycomers, which method comprises: (a) contacting the sample with
the microarray of claim 1 or 3, wherein each known glycomer is
affixed at at least one discrete locus, and wherein the contacting
is performed under conditions which would permit an agent, if
present in the sample, to specifically bind to its corresponding
glycomer in the microarray; (b) for each known glycomer in the
microarray, determining the amount of agent specifically bound
thereto; and (c) comparing the amounts so determined to a known
standard, thereby determining the amount of the one or more agents
in the sample.
54. The method of claim 53, wherein the agent is an antibody which
correlates with an inflammatory disease.
55. The method of claim 53, wherein the agent is an antibody which
correlates with an infection.
56. The method of claim 53, wherein the agent is an antibody which
correlates with the presence of a tumor.
57. The method of claim 53, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to a plurality of glycomers.
58. The method of claim 53, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to one glycomer.
59. A method of determining the amount of one or more agents in a
sample, each of which specifically binds to one or more known
insoluble proteins, which method comprises: (a) contacting the
sample with the microarray of claim 1 or 3, wherein each known
insoluble protein is affixed at at least one discrete locus, and
wherein the contacting is performed under conditions which would
permit an agent, if present in the sample, to specifically bind to
its corresponding insoluble protein in the microarray; (b) for each
known insoluble protein in the microarray, determining the amount
of agent specifically bound thereto; and (c) comparing the amounts
so determined to a known standard, thereby determining the amount
of the one or more agents in the sample.
60. The method of claim 59, wherein the agent is an antibody which
correlates with an inflammatory disease.
61. The method of claim 59, wherein the agent is an antibody which
correlates with an infection.
62. The method of claim 59, wherein the agent is an antibody which
correlates with the presence of a tumor.
63. The method of claim 59, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to a plurality of insoluble proteins.
64. The method of claim 59, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to one insoluble protein.
65. A method of determining the amount of one or more agents in a
sample, each of which specifically binds to one or more known
antibodies or lectins, which method comprises: (a) contacting the
sample with the microarray of claim 1 or 3, wherein each known
antibody or lectin is affixed at at least one discrete locus, and
wherein the contacting is performed under conditions which would
permit an agent, if present in the sample, to specifically bind to
its corresponding antibody or lectin in the microarray; (b) for
each known antibody or lectin in the microarray, determining the
amount of agent specifically bound thereto; and (c) comparing the
amounts so determined to a known standard, thereby determining the
amount of the one or more agents in the sample.
66. The method of claim 65, wherein the agent is an antibody which
correlates with an inflammatory disease.
67. The method of claim 65, wherein the agent is an antibody which
correlates with an infection.
68. The method of claim 65, wherein the agent is an antibody which
correlates with the presence of a tumor.
69. The method of claim 65, which method comprises determining the
amount of a plurality of agents in the sample, each of which binds
to a plurality of lectins or antibodies.
70. The method of claim 65, wherein the method comprises
determining the amount of a plurality of agents in the sample, each
of which binds to one lectin or antibody.
71. A method of determining whether a subject is afflicted with a
disorder characterized by the presence or absence in an afflicted
subject of an agent which specifically binds to a known glycomer,
which method comprises: (a) contacting a suitable sample from the
subject with the microarray of claim 1 or 3, wherein the known
glycomer is affixed at at least one discrete locus and wherein the
contacting is performed under conditions which would permit the
agent, if present in the sample, to specifically bind to the known
glycomer in the microarray; and (b) determining whether the known
glycomer in the microarray has the agent specifically bound
thereto, thereby determining whether the subject is afflicted with
the disorder.
72. The method of claim 71, wherein the subject is human.
73. The method of claim 71, wherein the disorder is an inflammatory
disorder.
74. The method of claim 73, wherein the inflammatory disorder is
celiac disease.
75. A method of determining whether a subject is afflicted with a
disorder characterized by the presence or absence in an afflicted
subject of an agent which specifically binds to a known insoluble
protein, which method comprises: (a) contacting a suitable sample
from the subject with the microarray of claim 1 or 3, wherein the
known insoluble protein is affixed at at least one discrete locus
and wherein the contacting is performed under conditions which
would permit the agent, if present in the sample, to specifically
bind to the known insoluble protein in the microarray; and (b)
determining whether the known insoluble protein in the microarray
has the agent specifically bound thereto, thereby determining
whether the subject is afflicted with the disorder.
76. The method of claim 75, wherein the subject is human.
77. A method of determining whether a subject is afflicted with a
disorder characterized by the presence or absence in an afflicted
subject of an agent which specifically binds to a known antibody or
lectin, which method comprises: (a) contacting a suitable sample
from the subject with the microarray of claim 1 or 3, wherein the
known antibody or lectin is affixed at at least one discrete locus
and wherein the contacting is performed under conditions which
would permit the agent, if present in the sample, to specifically
bind to the known antibody or lectin in the microarray; and (b)
determining whether the known antibody or lectin in the microarray
has the agent specifically bound thereto, thereby determining
whether the subject is afflicted with the disorder.
78. The method of claim 77, wherein the subject is human.
79. The method of claim 77, wherein the disorder is HIV-1
infection.
80. A method of determining whether an antibody known to
specifically bind to a first glycomer also specifically binds to a
second glycomer, which method comprises: (a) contacting the
antibody with the microarray of claim 1 or 3, wherein a plurality
of glycomers, other than the first glycomer, are affixed at
discrete loci in the microarray, and wherein the contacting is
performed under conditions which would permit the antibody to
specifically bind to the first glycomer if it were present in the
microarray; and (b) determining whether any of the glycomers in the
microarray, other than the first glycomer, has the antibody
specifically bound thereto, thereby determining whether the
antibody also specifically binds to a second glycomer.
81. A method of determining whether an antibody known to
specifically bind to a first insoluble protein also specifically
binds to a second insoluble protein, which method comprises: (a)
contacting the antibody with the microarray of claim 1 or 3,
wherein a plurality of insoluble proteins, other than the first
insoluble protein, are affixed at discrete loci in the microarray,
and wherein the contacting is performed under conditions which
would permit the antibody to specifically bind to the first
insoluble protein if it were present in the microarray; and (b)
determining whether any of the insoluble proteins in the
microarray, other than the first insoluble protein, has the
antibody specifically bound thereto, thereby determining whether
the antibody also specifically binds to a second insoluble
protein.
82. A method of making a microarray comprising a nitrocellulose or
Hydrogel support having affixed to its surface at discrete loci a
plurality of compounds, which method comprises contacting the
nitrocellulose or Hydrogel support with the compounds under
suitable conditions, whereby (a) at at least one discrete locus is
affixed a compound selected from the group consisting of a
glycomer, an insoluble protein, a lectin and an antibody, and (b)
the composition of compounds at each discrete locus differs from
the composition of compounds at at least one other discrete
locus.
83. A method of making a microarray comprising a plurality of
nitrocellulose or Hydrogel supports, each support having one or a
plurality of compounds affixed to its surface at a single discrete
locus or a plurality of compounds affixed to its surface at
discrete loci, which method comprises contacting the nitrocellulose
or Hydrogel supports with the compounds under suitable conditions,
whereby (a) at at least one discrete locus is affixed a compound
selected from the group consisting of a glycomer, an insoluble
protein, a lectin and an antibody, and (b) the composition of
compounds at each discrete locus differs from the composition of
compounds at at least one other discrete locus.
84. A method of making the article of claim 16 comprising
contacting a nitrocellulose or Hydrogel support with dextran at
discrete loci under suitable conditions.
85. The method of claim 84, further comprising the step of affixing
at least one compound to the dextran at each discrete locus,
whereby the composition of compounds at each discrete locus differs
from the composition of compounds at at least one other discrete
locus.
86. A method of making the article of claim 20 comprising
contacting a plurality of nitrocellulose or Hydrogel supports with
dextran, whereby each support has dextran affixed to its surface at
one or more discrete loci.
87. The method of claim 86, further comprising the step of affixing
at least one compound to the dextran at each discrete locus,
whereby the composition of compounds at each discrete locus differs
from the composition of compounds at at least one other discrete
locus.
88. A kit comprising the microarray of claim 1, 3, 18 or 22 and
instructions for use.
89. A kit comprising the microarray of claim 1, 3, 18 or 22 and a
desiccant.
90. A kit comprising the microarray of claim 1, 3, 18 or 22
immersed in an aqueous solution.
91. A kit for practicing the method of claim 71, which comprises:
(a) a microarray comprising a nitrocellulose or Hydrogel support
having affixed to its surface at discrete loci a plurality of
compounds, wherein (i) at at least one discrete locus is affixed
the glycomer to which the agent present or absent in an afflicted
subject specifically binds, and (ii) the composition of compounds
at each discrete locus differs from the composition of compounds at
at least one other discrete locus; and (b) instructions for
use.
92. A kit for practicing the method of claim 75, which comprises:
(a) a microarray comprising a nitrocellulose or Hydrogel support
having affixed to its surface at discrete loci a plurality of
compounds, wherein (i) at at least one discrete locus is affixed
the insoluble protein to which the agent present or absent in an
afflicted subject specifically binds, and (ii) the composition of
compounds at each discrete locus differs from the composition of
compounds at at least one other discrete locus; and (b)
instructions for use.
93. A kit for practicing the method of claim 77, which comprises:
(a) a microarray comprising a nitrocellulose or Hydrogel support
having affixed to its surface at discrete loci a plurality of
compounds, wherein (i) at at least one discrete locus is affixed
the antibody or lectin to which the agent present or absent in an
afflicted subject specifically binds, and (ii) the composition of
compounds at each discrete locus differs from the composition of
compounds at at least one other discrete locus; and (b)
instructions for use.
94. An antibody capable of specifically binding to a glycomer
present on the surface of a mammalian macrophage, which glycomer,
or structural mimic thereof, is also endogenous to, and present on
the surface of, a bacterial cell.
95. The antibody of claim 94, wherein the antibody is a groove-type
antibody.
96. The antibody of claim 94, wherein the antibody is designated
4.3.F1 (ATCC Accession No. PTA-3259).
97. The antibody of claim 94, wherein the antibody is designated
45.21.1 (ATCC Accession No. PTA-3260).
98. An antibody capable of specifically binding to a glycomer
present on the surface of a mammalian intestinal epithelial cell,
which glycomer, or structural mimic thereof, is also endogenous to,
and present on the surface of, a bacterial cell.
99. The antibody of claim 98, wherein the antibody is a cavity-type
antibody.
100. The antibody of claim 98, wherein the antibody is designated
16.4.12E (ATCC Accession No. PTA-3261).
101. A method of determining whether a subject is afflicted with a
disorder characterized by the presence of a glycomer on the surface
of macrophages in an afflicted subject, which glycomer, or
structural mimic thereof, is also endogenous to, and present on the
surface of, a bacterial cell, comprising: (a) contacting a sample
of the subject's macrophages with the antibody of claim 94; and (b)
determining whether the antibody specifically binds to the
macrophages in the sample, such binding indicating that the subject
is afflicted with the disorder.
102. The method of claim 101, wherein the subject is human.
103. The method of claim 101, wherein the disorder is an immune
disorder or an inflammatory disorder.
104. A method of determining whether a subject is afflicted with a
disorder characterized by the presence of a glycomer on the surface
of intestinal epithelial cells in an afflicted subject, which
glycomer, or structural mimic thereof, is also endogenous to, and
present on the surface of, a bacterial cell, comprising: (a)
contacting a sample of the subject's intestinal epithelial cells
with the antibody of claim 98; and (b) determining whether the
antibody specifically binds to the intestinal epithelial cells in
the sample, such binding indicating that the subject is afflicted
with the disorder.
105. The method of claim 104, wherein the subject is human.
106. The method of claim 104, wherein the disorder is an immune
disorder or an inflammatory disorder.
107. The method of claim 106, wherein the disorder is celiac
disease.
Description
[0001] This invention is a continuation-in-part and claims the
benefit of PCT International Application No. PCT/US02/11612, filed
Apr. 10, 2002, which is a continuation-in-part and claims the
benefit of U.S. Provisional Application No. 60/282,926, filed Apr.
10, 2001, the contents of which are hereby incorporated by
reference into this application.
[0002] Throughout this application, various references are cited.
Disclosure of these references in their entirety is hereby
incorporated by reference into this application to more fully
describe the state of the art to which this invention pertains.
[0003] The invention described herein was made with Government
support under grant number AI45326 from the National Institutes of
Health. Accordingly, the United States Government has certain
rights in this invention.
BACKGROUND OF THE INVENTION
[0004] Genomics
[0005] The Human Genome Project is rapidly approaching its end: the
complete mapping and sequencing of the human genome, and the
identification of all genes therein. Emerging from this effort is a
new generation of biotechnologies, collectively known as
"functional genomics". These technologies, including DNA chips (1)
and cDNA microarrays (2,3), make use of the sequence information
and genetic materials provided by the human genome project, combine
advanced laser and fluorescence sensor technology, and take
advantage of computer-aided large-scale data management systems.
Differing from classical molecular biology methods which focus on a
specific gene or its product, these new approaches monitor the
expression of genes on a genome-wide scale, and identify their
characteristic overall patterns. The scope of biological
investigation has therefore been expanded from the study of a
single gene or protein to the study of numerous genes and/or
proteins simultaneously.
[0006] Proteomics
[0007] Proteins are the final gene products, acting as fundamental
elements of living organisms. However, the amount of mRNA
expression does not always indicate the level of its encoded
protein in a cell. The protein molecule has its own life span and
kinetics of metabolism. There are specialized cellular machineries,
such as the ubiquitin-dependent and -independent pathways of
protein degradation, allowing rapid turnover of a protein when its
function is no longer required. The fate of a newly synthesized
protein is also significantly influenced by post-translational
modifications, such as phosphorylation, glycosylation, acetylation
or myristylation, at specific amino acid residues. Such molecular
modifications are frequently regulated differentially and/or
developmentally, establishing a specific function, or playing a
structural role, for a given protein. Thus, it has been generally
accepted that a better understanding of the genome's function will
not be possible without protein analysis. Developing technologies
for a genome-wide analysis of protein expression and
post-translational modification represents a major challenge to the
scientific community (4,5).
[0008] Glycomics
[0009] Carbohydrate-containing macromolecules are the secondary
products of genes. Their synthesis requires multiple enzymatic
reactions and many steps of intracellular trafficking,
transportation and modification. Multiple genes contribute to the
synthesis of cellular elements containing complex carbohydrates.
"Glycomics", a new scientific discipline, has emerged to create a
comprehensive understanding of the structure, function, synthesis
and genetic regulation of cellular carbohydrate molecules.
[0010] Carbohydrates are abundant on cell surfaces, existing as
either membrane-bound glycoconjugates or secreted substances. These
molecules play fundamental structural and protective roles. They
are also abundant intracellularly, and serve as an active and
dynamic energy reservoir.
[0011] Recent studies further demonstrated that many important
signaling and regulatory processes are mediated by the interaction
of carbohydrate-ligands and their receptors (6,7). Abnormal
expression of carbohydrate moieties may occur in cells that are
undergoing malignant transformation. These moieties may therefore
serve as molecular targets for tumor diagnosis or therapy.
[0012] The carbohydrate molecules of microorganisms are important
in establishing the biological relationships of microbes and their
hosts (7-9). These relationships especially include the host
recognition of microorganisms and the induction of an immune
response by a microbial antigen. The carbohydrate moieties of
microbial antigens frequently serve as the key structures for
immune recognition (10). Identifying such determinants is of
fundamental importance for understanding the molecular mechanisms
of host recognition and immune responses.
[0013] Existing Technologies
[0014] Technologies suitable for monitoring protein expression on a
genome-wide scale and for characterizing a wide range of
ligand-receptor interactions such as protein-protein reactions,
carbohydrate-protein reactions and the interaction of synthetic
small molecules and cellular components have yet to be developed.
Current methods for specifically detecting and quantifying a
protein or a microbial polysaccharide include antigen/antibody
based-immunoassays. These assays include (a) classical direct
immunoassays, such as immunodiffusion, immunoelectrophoresis,
agglutination and immunoprecipitation assays, and (b) recently
developed methods such as immunofluorescence, radioimmunoassay
(RIA), enzyme-immunoassay (EIA) and western blot assays. These
approaches exploit the specificity of antigen-antibody
interactions. However, they are designed for analyzing only one
agent at a time, and are therefore limited as to the number of
molecules that can be analyzed in a single assay.
[0015] In sum, a single technology useful for the simultaneous
study of numerous molecules, be they protein, carbohydrate or
combinations thereof, is sorely needed to advance both proteomics
and glycomics.
SUMMARY OF THE INVENTION
[0016] This invention provides four microarrays and two articles
useful for making same. The first microarray comprises a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (a) at at least one
discrete locus is affixed a compound selected from the group
consisting of a glycomer, an insoluble protein, a lectin and an
antibody, and (b) the composition of compounds at each discrete
locus differs from the composition of compounds at at least one
other discrete locus.
[0017] The second microarray comprises a plurality of
nitrocellulose or Hydrogel supports, each support having one or a
plurality of compounds affixed to its surface at a single discrete
locus or a plurality of compounds affixed to its surface at
discrete loci, wherein (a) at at least one discrete locus is
affixed a compound selected from the group consisting of a
glycomer, an insoluble protein, a lectin and an antibody, and (b)
the composition of compounds at each discrete locus differs from
the composition of compounds at at least one other discrete
locus.
[0018] The first article comprises a nitrocellulose or Hydrogel
support having dextran affixed to its surface at discrete loci. In
one embodiment the dextran is .alpha.(1,6) dextran.
[0019] The third microarray comprises the first article, wherein at
least one compound is affixed to the dextran at each discrete
locus, the composition of compounds at each discrete locus
differing from the composition of compounds at at least one other
discrete locus.
[0020] The second article comprises a plurality of nitrocellulose
or Hydrogel supports, each support having dextran affixed to its
surface at one or more discrete loci. In one embodiment the dextran
is .alpha.(1,6) dextran.
[0021] The fourth microarray comprises the second article, wherein
at least one compound is affixed to the dextran at each discrete
locus, the composition of compounds at each discrete locus
differing from the composition of compounds at at least one other
discrete locus.
[0022] This invention provides three methods for detecting the
presence of agents in a sample. The first method is a method of
detecting in a sample the presence of one or more agents which
specifically bind to one or more known glycomers, which method
comprises: (a) contacting the sample with the first or second
microarray, wherein each known glycomer is affixed at at least one
discrete locus and wherein the contacting is performed under
conditions which would permit an agent, if present in the sample,
to specifically bind to its corresponding glycomer in the
microarray; and (b) determining whether any known glycomer in the
microarray has an agent specifically bound thereto, thereby
detecting the presence of the one or more agents in the sample.
[0023] The second method is a method of detecting in a sample the
presence of one or more agents which specifically bind to one or
more known insoluble proteins, which method comprises: (a)
contacting the sample with the first or second microarray, wherein
each known insoluble protein is affixed at at least one discrete
locus and wherein the contacting is performed under conditions
which would permit an agent, if present in the sample, to
specifically bind to its corresponding insoluble protein in the
microarray; and (b) determining whether any known insoluble protein
in the microarray has an agent specifically bound thereto, thereby
detecting the presence of the one or more agents in the sample.
[0024] The third method is a method of detecting in a sample the
presence of one or more agents which specifically bind to one or
more known antibodies or lectins, which method comprises: (a)
contacting the sample with the first or second microarray, wherein
each known antibody or lectin is affixed at at least one discrete
locus and wherein the contacting is performed under conditions
which would permit an agent, if present in the sample, to
specifically bind to its corresponding antibody or lectin in the
microarray; and (b) determining whether any known antibody or
lectin in the microarray has an agent specifically bound thereto,
thereby detecting the presence of the one or more agents in the
sample.
[0025] This invention further provides three quantitative methods.
The first method is a method of determining the amount of one or
more agents in a sample, each of which specifically binds to one or
more known glycomers, which method comprises: (a) contacting the
sample with the first or second microarray, wherein each known
glycomer is affixed at at least one discrete locus, and wherein the
contacting is performed under conditions which would permit an
agent, if present in the sample, to specifically bind to its
corresponding glycomer in the microarray; (b) for each known
glycomer in the microarray, determining the amount of agent
specifically bound thereto; and (c) comparing the amounts so
determined to a known standard, thereby determining the amount of
the one or more agents in the sample.
[0026] The second method is a method of determining the amount of
one or more agents in a sample, each of which specifically binds to
one or more known insoluble proteins, which method comprises: (a)
contacting the sample with the first or second microarray, wherein
each known insoluble protein is affixed at at least one discrete
locus, and wherein the contacting is performed under conditions
which would permit an agent, if present in the sample, to
specifically bind to its corresponding insoluble protein in the
microarray; (b) for each known insoluble protein in the microarray,
determining the amount of agent specifically bound thereto; and (c)
comparing the amounts so determined to a known standard, thereby
determining the amount of the one or more agents in the sample.
[0027] The third method is a method of determining the amount of
one or more agents in a sample, each of which specifically binds to
one or more known antibodies or lectins, which method comprises:
(a) contacting the sample with the first or second microarray,
wherein each known antibody or lectin is affixed at at least one
discrete locus, and wherein the contacting is performed under
conditions which would permit an agent, if present in the sample,
to specifically bind to its corresponding antibody or lectin in the
microarray; (b) for each known antibody or lectin in the
microarray, determining the amount of agent specifically bound
thereto; and (c) comparing the amounts so determined to a known
standard, thereby determining the amount of the one or more agents
in the sample.
[0028] This invention further provides three diagnostic methods.
The first method is a method of determining whether a subject is
afflicted with a disorder characterized by the presence or absence
in an afflicted subject of an agent which specifically binds to a
known glycomer, which method comprises: (a) contacting a suitable
sample from the subject with the first or second microarray,
wherein the known glycomer is affixed at at least one discrete
locus and wherein the contacting is performed under conditions
which would permit the agent, if present in the sample, to
specifically bind to the known glycomer in the microarray; and (b)
determining whether the known glycomer in the microarray has the
agent specifically bound thereto, thereby determining whether the
subject is afflicted with the disorder.
[0029] The second method is a method of determining whether a
subject is afflicted with a disorder characterized by the presence
or absence in an afflicted subject of an agent which specifically
binds to a known insoluble protein, which method comprises: (a)
contacting a suitable sample from the subject with the first or
second microarray, wherein the known insoluble protein is affixed
at at least one discrete locus and wherein the contacting is
performed under conditions which would permit the agent, if present
in the sample, to specifically bind to the known insoluble protein
in the microarray; and (b) determining whether the known insoluble
protein in the microarray has the agent specifically bound thereto,
thereby determining whether the subject is afflicted with the
disorder.
[0030] The third method is a method of determining whether a
subject is afflicted with a disorder characterized by the presence
or absence in an afflicted subject of an agent which specifically
binds to a known antibody or lectin, which method comprises: (a)
contacting a suitable sample from the subject with the first or
second microarray, wherein the known antibody or lectin is affixed
at at least one discrete locus and wherein the contacting is
performed under conditions which would permit the agent, if present
in the sample, to specifically bind to the known antibody or lectin
in the microarray; and (b) determining whether the known antibody
or lectin in the microarray has the agent specifically bound
thereto, thereby determining whether the subject is afflicted with
the disorder.
[0031] This invention further provides a method of determining
whether an antibody known to specifically bind to a first glycomer
also specifically binds to a second glycomer, which method
comprises: (a) contacting the antibody with the first or second
microarray, wherein a plurality of glycomers, other than the first
glycomer, are affixed at discrete loci in the microarray, and
wherein the contacting is performed under conditions which would
permit the antibody to specifically bind to the first glycomer if
it were present in the microarray; and (b) determining whether any
of the glycomers in the microarray, other than the first glycomer,
has the antibody specifically bound thereto, thereby determining
whether the antibody also specifically binds to a second
glycomer.
[0032] This invention further provides a method of determining
whether an antibody known to specifically bind to a first insoluble
protein also specifically binds to a second insoluble protein,
which method comprises: (a) contacting the antibody with the first
or second microarray, wherein a plurality of insoluble proteins,
other than the first insoluble protein, are affixed at discrete
loci in the microarray, and wherein the contacting is performed
under conditions which would permit the antibody to specifically
bind to the first insoluble protein if it were present in the
microarray; and (b) determining whether any of the insoluble
proteins in the microarray, other than the first insoluble protein,
has the antibody specifically bound thereto, thereby determining
whether the antibody also specifically binds to a second insoluble
protein.
[0033] This invention further provides a method of making a
microarray comprising a nitrocellulose or Hydrogel support having
affixed to its surface at discrete loci a plurality of compounds,
which method comprises contacting the nitrocellulose or Hydrogel
support with the compounds under suitable conditions, whereby (a)
at at least one discrete locus is affixed a compound selected from
the group consisting of a glycomer, an insoluble protein, a lectin
and an antibody, and (b) the composition of compounds at each
discrete locus differs from the composition of compounds at at
least one other discrete locus.
[0034] This invention further provides a method of making a
microarray comprising a plurality of nitrocellulose or Hydrogel
supports, each support having one or a plurality of compounds
affixed to its surface at a single discrete locus or a plurality of
compounds affixed to its surface at discrete loci, which method
comprises contacting the nitrocellulose or Hydrogel supports with
the compounds under suitable conditions, whereby (a) at at least
one discrete locus is affixed a compound selected from the group
consisting of a glycomer, an insoluble protein, a lectin and an
antibody, and (b) the composition of compounds at each discrete
locus differs from the composition of compounds at at least one
other discrete locus.
[0035] This invention further provides a method of making the first
article comprising contacting a nitrocellulose or Hydrogel support
with dextran at discrete loci under suitable conditions.
[0036] This invention further provides a method of making the
second article comprising contacting a plurality of nitrocellulose
or Hydrogel supports with dextran, whereby each support has dextran
affixed to its surface at one or more discrete loci.
[0037] This invention further provides six kits. The first kit
comprises one of the instant microarrays and instructions for use.
The second kit comprises one of the instant microarrays and a
desiccant. The third kit comprises one of the instant microarrays
immersed in an aqueous solution.
[0038] The fourth kit is a kit for practicing the first diagnostic
method, which comprises: (a) a microarray comprising a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (i) at at least one
discrete locus is affixed the glycomer to which the agent present
or absent in an afflicted subject specifically binds, and (ii) the
composition of compounds at each discrete locus differs from the
composition of compounds at at least one other discrete locus; and
(b) instructions for use.
[0039] The fifth kit is a kit for practicing the second diagnostic
method, which comprises: (a) a microarray comprising a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (i) at at least one
discrete locus is affixed the insoluble protein to which the agent
present or absent in an afflicted subject specifically binds, and
(ii) the composition of compounds at each discrete locus differs
from the composition of compounds at at least one other discrete
locus; and (b) instructions for use.
[0040] The sixth kit is a kit for practicing the third diagnostic
method, which comprises: (a) a microarray comprising a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (i) at at least one
discrete locus is affixed the antibody or lectin to which the agent
present or absent in an afflicted subject specifically binds, and
(ii) the composition of compounds at each discrete locus differs
from the composition of compounds at at least one other discrete
locus; and (b) instructions for use.
[0041] This invention further provides a first antibody capable of
specifically binding to a glycomer present on the surface of a
mammalian macrophage, which glycomer, or structural mimic thereof,
is also endogenous to, and present on the surface of, a bacterial
cell.
[0042] This invention further provides a second antibody capable of
specifically binding to a glycomer present on the surface of a
mammalian intestinal epithelial cell, which glycomer, or structural
mimic thereof, is also endogenous to, and present on the surface
of, a bacterial cell.
[0043] This invention further provides a method of determining
whether a subject is afflicted with a disorder characterized by the
presence of a glycomer on the surface of macrophages in an
afflicted subject, which glycomer, or structural mimic thereof, is
also endogenous to, and present on the surface of, a bacterial
cell, comprising: (a) contacting a sample of the subject's
macrophages with the first antibody; and (b) determining whether
the antibody specifically binds to the macrophages in the sample,
such binding indicating that the subject is afflicted with the
disorder.
[0044] Finally, this invention provides a method of determining
whether a subject is afflicted with a disorder characterized by the
presence of a glycomer on the surface of intestinal epithelial
cells in an afflicted subject, which glycomer, or structural mimic
thereof, is also endogenous to, and present on the surface of, a
bacterial cell, comprising: (a) contacting a sample of the
subject's intestinal epithelial cells with the second antibody; and
(b) determining whether the antibody specifically binds to the
intestinal epithelial cells in the sample, such binding indicating
that the subject is afflicted with the disorder.
BRIEF DESCRIPTION OF THE FIGURES
[0045] FIG. 1
[0046] This Figure shows a carbohydrate microarray and its
application in characterizing the epitope-binding specificity of
monoclonal antibodies ("mAb"). Dextran preparations of defined
structural characteristics, including N279, LD7, B1299S and B1355S,
were immobilized on a nitrocellulose-coated micro-glass slide in
serial dilutions and stained with anti-.alpha.(1,6)dextran
antibodies. These antibodies were either a groove-type antibody,
i.e., 4.3.F1, or a cavity-type antibody, i.e., 16.4.12E, and were
conjugated with fluorescence. Their distinct epitope-binding
specificities were visualized by scanning the carbohydrate
microarray using a GMS 418 microarray scanner.
[0047] FIG. 2
[0048] This Figure shows an antigen-based microarray and its
application in studying the cross-reactivity of monoclonal
antibodies. Forty-nine distinct antigen preparations, including
microbial polysaccharides, blood group substances and other
glycoconjugates, were arrayed on slides and incubated with
fluorescence-labeled mabs. FIG. 2A: anti-DEX 4.3.F1; FIG. 2B:
anti-DEX 16.4.12E. Intensity values of cross-reacting spots were
compared with those of specific binding to .alpha.(1,6)dextran
N279. N279 was applied as a series of 1:5 dilutions of a 100
.mu.g/ml solution (a). Other antigens were applied as 500 .mu.g/ml.
Identical antigens are arrayed in FIG. 2A and FIG. 2B.
[0049] FIG. 3
[0050] This Figure shows the detection of a cell population in the
small intestine of an adult mouse by a groove-type
anti-.alpha.(1,6)dextran antibody 4.3.F1 (IgG3), which showed
cross-reactivity to a chondroitin sulfate B preparation. The
cryostat sections of small intestine were stained either by mAb
4.3.F1 (IgG3) or by an IgG3 isotype control mAb obtained from BD
PharMingen. The two mAbs were fluorescent conjugates. Sections were
co-stained with DAPI for the cell nucleus to visualize the overall
tissue structure. The 4.3.F1-positive cells were seen in the lamina
propria of the small intestine. FIGS. 3A-3D: mAb 4.3.F1; FIGS.
3E-3H: isotype control.
[0051] FIG. 4
[0052] This Figure shows that groove-type and cavity-type
anti-.alpha.(1,6)dextran monoclonal antibodies recognize distinct
cellular markers: a groove-type mAb 45.21.1 (IgA) identifies a cell
population in the lamina propria of the small intestine (FIGS. 4C
and 4F), and a cavity-type mAb 16.4.12E (IgA) stains the epithelial
cells in the crypts of the small intestine (FIGS. 4B and 4E). An
IgA isotype control mAb purchased from BD PharMingen was applied as
a background control (FIGS. 4A and 4D).
[0053] FIG. 5
[0054] This Figure shows the recognition of a cell population in
the human small intestinal tissue using anti-.alpha.(1,6)dextran
antibodies. The intestinal section of a normal (FIGS. 5A and 5B)
and of a celiac individual (FIGS. 5C and 5D) were stained with the
fluorescence-conjugate of mAb 16.4.12E (FIGS. 5B and 5D) and
co-stained with DAPI to reveal the intestinal structures (FIGS. 5A
and 5C).
[0055] FIG. 6
[0056] This Figure shows the immobilization of polysaccharides on a
nitrocellulose-coated glass slide. Panel A: Image of the
carbohydrate microarray (microarrays of dextrans and inulin) spots
before and after washing. Panel B: Quantitative illustration of the
relation of fluorescence intensity and the concentration of printed
carbohydrate microarrays before and after washing. Fluorescent
conjugates of dextrans or inulin were dissolved in saline (0.9%
NaCl) and spotted at an initial concentration of 10 mg/ml and then
diluted in serial dilutions of 1:5. The microarray slides were
scanned before and after washing. The data of six repeats of the
same experiment is statistically analyzed and presented. Legend:
-.box-solid.-: 2000 k; -.tangle-solidup.-: 70 k; -.circle-solid.-:
20 k; -.diamond-solid.-: Inulin.
[0057] FIG. 7
[0058] This Figure shows the immunological characterization of
surface-immobilized dextran molecules. Panel A: Microarray binding
curves of a groove-type anti-Dex 4.3F1 (IgG3/Kappa) and a
cavity-type anti-dextran 16.4.12E (IgA/Kappa) to dextran molecules
of distinct structure. Dextran molecules were printed with an
initial concentration of 0.1 mg/ml and diluted by a 1:5 series
titration. The printed arrays were washed to remove unbound
antigens and then stained with biotinylated anti-dextran, either
4.3F1 or 16.4.12E, at a concentration of 1 .mu.g/ml, and then
stained with Cy3-streptavidin at 1:500 dilutions. The readout of
the experiment (i.e., fluorescent intensity of the microspot)
reflects the amounts of antigen immobilized and epitopes displayed
for antibody recognition. The cavity-type mAb 16.4.12E bound to
N279 and B12995, but not LD7. By contrast, the groove-type mAb
4.3F1 bound to the dextran preparations N279 and LD7, but bound
poorly to B1299S. Panel B: ELISA binding curve of anti-Dex 4.3F1
and 16.4.12E. Dextran preparations were coated on an ELISA plate at
an initial concentration of 10 .mu.g/ml and then diluted by a 1:5
series titration in 0.02 M borate-buffered saline, pH 8.0. The
antigen-coated plates were incubated with biotinylated
anti-dextrans at a concentration of 1 .mu.g/ml. The bound
antibodies were revealed with an alkaline phosphatase
(AP)-streptavidin conjugate and AP substrate. Legend:
-.diamond-solid.-: N279; -.quadrature.-: LD7; -.tangle-solidup.-:
B-1299S. Left Panel: 4.3F1 (Groove-type); Right Panel: 16.4.12
(Cavity-Type).
[0059] FIG. 8
[0060] Bacillus anthracis exposes and releases a number of antigens
of distinct structural characteristics to trigger and induce a
comprehensive picture of a host response. Left: Schematic of the
life cycle of Bacillus anthracis. Dormant spores present in vitro
are highly resistant to adverse environmental conditions. In a
suitable environment, spores establish vegetative growth. In an
early infection, these infective particles are ingested by the
phagocytic cells and accumulate in the local lymphoid tissue. Some
may survive from the phagocytes and initiate their germination and
vegetative growth. The vegetative form of the bacteria is
square-ended and capsulated. In the late infection, they multiply
rapidly, express their virulence factors to kill the host and
develop to the stage of resporulation in vivo. Antigens &
toxins: Vegetative bacillus releases multiple factors, such as
toxins, protein factors, and soluble polysaccharides (1-3,4 of the
Third Series of Experiments). The protein fractions, such as the
protective antigen, named PA, can provide strong protection to the
immunized animals. It is now well understood that PA is an
integrated component of the lethal toxin of Bacillus anthracis. It
binds to a specific cellular receptor and forms toxic, cell bound
complexes with edema factor (EF) and lethal factor (LF) (1-3 of the
Third Series of Experiments). Neutralization antibodies to PA or a
polyvalent factor that inhibits the formation of the complex may
protect animals from the lethal attack by the toxin (5 of the Third
Series of Experiments). A considerable amount of polysaccharides
are also present in the culture media of the growing bacteria. Its
sugar compositions are similar (if not identical) to the cell wall
Gal-NAG polysaccharide. Right: An outline of the human immune
system. The native immunity forms the first line of a host
anti-infection response. These include macrophages, natural killer
cells (NK), pre-existing "natural antibody" of IgM isotype and
perhaps a specific B cell lineage, the B-1 cells, TCR
.gamma..delta. T cells, and other cells. In the anthrax infection,
the phagocytic cells may play multiple roles in the host-microbe
interaction. These may include both protective and pathogenic
effects (see below for details). The acquired immune system
includes B cells (the bone marrow derived B cells, or B-2 cells)
and T cells (the thymus derived TCR .alpha..beta. T cells). B cells
mount a specific antibody response to a microbial antigen, either a
T-independent antigen, such as an anthrax polysaccharide, or a
T-dependent antigen, for example the protective antigen (PA) of B.
anthracis; specific T cells can be activated by a TD protein
antigen to regulate a B cell responses, either positively (T
helper, Th1 and Th2) or negatively. There is also activation of
specific cytotoxic T cells (Tc), which can kill the cells that
express a foreign antigen. Many host cells, including immune cells
and non-immune cell types, may produce cytokines or other
inflammation factors to assist a host anti-infection response.
[0061] FIG. 9
[0062] This Figure shows a simple and efficient procedure for
producing a carbohydrate microarray. Microspotting: Carbohydrate
antigens were printed using Cartesian Technologies' PIXSYS 5500C
(Irvine, Calif.) with STEALTH 3 pins. Supporting substrate: FAST
Slides (Industrial partner A, Schleicher & Schuell, Keene,
N.H.). The printed carbohydrate microarrays were air dried and
stored at room temperature without desiccant before application.
Immuno-staining: Immediately before use, the microarrays were
rinsed with phosphate-buffered saline (PBS). The staining procedure
utilized is essentially identical to regular immunoflourescent
staining of tissue sections. Microarray-scanning: A ScanArray 5000
Standard Biochip Scanning System and its QuantArray software
(Packard Biochip Technologies, Inc.) were applied for scanning and
data capturing.
[0063] FIG. 10
[0064] This Figure shows a schematic of the 8-chamber
sub-arrays.
[0065] FIG. 11
[0066] This Figure shows probing of the repertoires of human serum
antibodies using Antigen Chip 4000. Left: HIV negative normal
serum. Right: Serum of an HIV-1 infected individual. For each
microarry analysis, 10 ml of serum were applied on an antigen chip
at 1:10 dilutions. Anti-human antibodies with distinct fluorescent
tags were applied to recognize and quantify the bound human IgG,
IgM and IgA. In this Figure, human IgG was stained in Red/Cy5 and
human IgM in Green/Cy3. The two images of contrasting colors were
overlaid. IgA human antibodies were detected on the same chip with
an anti-human IgA.sup.FITC (data was not shown)
[0067] FIG. 12
[0068] This Figure shows the scanning of human antibodies specific
for a large panel of HIV proteins using a protein-based microarray
biochip. Serum specimens of four normal individuals and six AIDS
patients were characterized by a protein biochip that displays a
large panel of HIV-1 proteins. Each preparation was printed four
times on the same biochip. For each assay, 10 .mu.l of serum were
applied at 1:10 dilutions on a single chip. Human IgG that was
captured by the immobilized antigens was recognized and quantified
by a Cy3-labeled second antibody. Data of each group, normal and
HIV-infected individuals, were statistically analyzed. Results were
presented as the mean value of the ratio of fluorescent intensity
over the background of given microspots (Histogram). Their standard
division was also shown. Significant variations that were observed
in the HIV-1 infected group may reflect the diversity of the HIV-1
specific antibody responses, as well as the level of antigenic
cross-reactivities of HIV-1 proteins that were expressed by
different clades or strains of HIV-1 virus.
[0069] FIG. 13
[0070] This Figure shows a schematic of a hypothetical structural
and immunological relationships of the type II backbone structure
of blood group substances, type XIV pneumococcal polysaccharide and
the cell wall Gal-NAG polysaccharide.
[0071] FIG. 14
[0072] This Figure shows a carbohydrate microarray characterization
of human and murine antibodies. Forty-eight distinct antigen
preparations were arrayed on slides at antigen concentrations of
0.5 mg/ml and 0.02 mg/ml. They were incubated with combined human
serum specimens at a concentration equivalent to 1:100 dilutions of
each specimen or with binotinylated mouse monoclonal antibodies at
1 mg/ml. The human IgM captured by microarrays was visualized using
an anti-human IgM-AP conjugate and the color developed using Vector
Red. The human IgG anti-carbohydrates were detected using a
biotinylated anti-human IgG. A Cy3-Streptoavidin conjugate was then
applied to visualize the human IgG or murine monoclonal antibodies
bound on microarray. The readout of the experiment, i.e.,
fluorescent intensity of the microspot, reflects the amounts of
antigen immobilized and epitopes displayed for antibody
recognition. Data of four repeats of microarray staining are
summarized in Table 2.
[0073] FIG. 15
[0074] This Figure shows the prediction of protein structure using
TMHMM version 2.0 (55 of the Third Series of Experiments): PX01-54
of B. anthracis encodes a S-layer protein, a novel molecular target
for anthrax diagnosis and vaccination.
[0075] FIG. 16
[0076] This Figure shows biochip detection of human antibody
reactivities to either anthrax polysaccharide or Pneumococcus type
XIV polysaccharide in mixed human serum specimen which confirms
that these antigen preparations are applicable for producing
diagnostic microarrays.
[0077] FIG. 17
[0078] This Figure shows a schematic of diagnosis and surveillance
of emerging infectious diseases using antigen arrays. An emerging
infection may attack human as well as animal populations. Detecting
specific antibodies in their bodily fluids using antigen
microarrays is a powerful means for the diagnosis and surveillance
of emerging infectious diseases. Left: schematic of the life cycle
of Bacillus anthracis; Right: host antibodies elicited by an
infection are important molecular targets for the diagnosis and
surveillance of emerging infectious diseases.
[0079] FIG. 18
[0080] This Figure shows long-lasting antigenic reactivities of
protein microarrays in some but not all preparations.
[0081] FIG. 19
[0082] This Figure shows that clustering analysis identified
reproducible global pattern of antigenic reactivities on
biochips.
[0083] FIG. 20
[0084] This Figure shows that purified GFPuv protein is stably
immobilized on the nitrocellulose-coated slide (upper row) but is
less stable on a poly-Lys treated DNA ready slide (bottom row). A
preparation of purified GFPuv was spotted on slide at 0.5 mg/ml
(left side) and 0.05 mg/ml (right side). Scanning was performed 1)
at the same day after spotting; 2) after overnight air-drying the
slide; and 3) after washing the slide extensively in a buffer
containing 0.1% Tween 20 1.times. PBS.
[0085] FIG. 21
[0086] This Figure shows the structure of PA and its four domains:
PA is a long, flat molecule of dimensions
100.times..sup..about.50.times.30 .ANG.. and is composed of four
domains.sup.46. Domain 1 (residues 1-258) comprises a b-sandwich
with jelly-roll topology, several small helices, and a pair of
adjacent calcium ions coordinated by residues in a variant of the
EF-hand motif. Domain 2 (residues 259-487) has a b-barrel core with
modified Greek-key topology and elaborate excursions, including a
large flexible loop between strands 2b2 and 2b3 that is implicated
in membrane insertion. Domain 3 (residues 488-595) has a
four-stranded mixed b-sheet, two smaller sheets and four helices;
it adopts the same fold as ferredoxins and resembles domain A of
toxic-shock-syndrome toxin-1. Domain 4 (residues 596-735) has an
initial hairpin and helix, followed by a b-sandwich with an
immunoglobulin-like fold. Domains 1, 2 and 3 are intimately
associated, but domain 4 has limited contact with the other 3
domains. D1 is the N-terminal domain, contains two calcium ions and
the cleavage site for activating proteases. D2 contains a large
flexible loop for membrane insertion. D3 is a small domain of
unknown function. D4 is the C-terminal receptor-binding
domain.(Domain description from Petosa et al. Nature Vol.
305(1997); image generated from the structure of protective antigen
using NCBI's Cn3D version 4.0).
[0087] FIG. 22
[0088] This Figure shows a schematic of a molecular approach to
construct GFP fusion proteins of PA and its derivatives. Primers
for PCR reactions are designed based on sequence information of
each domain target. Restriction sites are selected to avoid
digesting the PA chains or its domains. Following are primer
sequences of PA derived portion: Integral PA (nt bases
143,799-146,073): 5' ATGGTTCTTTAGCTTTCTG (19 mer, Start: 143,409);
3' CCTAGAATTACCTTATCCTATC (22 mer, Stop: 146,085). P63 (nt bases
144,376-146,070): 5' GCGAAGTACAAGTGCTGG (18 mer, Start: 144,363);
3' TTGAATGTGCAATTGTCCTC (20 mer, Stop: 146,321). PA Domain 1 (nt
bases 143,866-144,639): 5' TCAGGCAGAAGTTAAACAGG (20 mer, Start:
143,859); 3' GATAAGCTGCCACAAGGG (18 mer, Stop: 144,643). PA Domain
2 (nt bases 144,640-145,326): 5' TGGCAGCTTATCCGATTG (18 mer, Start:
144,632); 3' GCAGTTGTTTCTTGAATTTGCG (22 mer, Stop: 145,331). PA
Domain 3 (nt bases 145,327-145,650): 5.degree.
CAAGAAACAACTGCACGTATC (21 mer, Start: 145,318); 3.degree.
CTTCTCTATGAGCCTCCTTAAC (22 mer, Stop: 145,720). PA Domain 4 (nt
bases 145,651-146,070): 5' TATCAAGAATCA GTTAGCG (19 mer, Start:
145,548); 3' ACCTTATCCTATCTCATAGCC (21 mer, Stop: 146,076).
[0089] FIG. 23A
[0090] This Figure shows lectin staining of .alpha.-Gal specific
oligosaccharides displayed by the neoglycocojugates.
[0091] FIG. 23B
[0092] This Figure shows that .alpha.-Gal-sugar chains are
displayed by the PAGE-based neoglycocojugates and are detected by
the human anti-.alpha.-Gal antibodies.
DETAILED DESCRIPTION OF THE INVENTION
[0093] Definitions
[0094] As used in this application, except as otherwise expressly
provided herein, each of the following terms shall have the meaning
set forth below.
[0095] "Affixed" shall mean attached by any means. In one
embodiment, affixed shall mean attached by a covalent bond. In
another embodiment, affixed shall mean attached non-covalently.
[0096] "Agent" shall mean any chemical entity, including, without
limitation, a glycomer, a protein, an antibody, a lectin, a nucleic
acid, a small molecule, and any combination thereof.
[0097] "Antibody" shall mean (a) an immunoglobulin molecule
comprising two heavy chains and two light chains and which
recognizes an antigen; (b) polyclonal and monoclonal immunoglobulin
molecules; and (c) monovalent and divalent fragments thereof.
Immunoglobulin molecules may derive from any of the commonly known
classes, including but not limited to IgA, secretory IgA, IgG and
IgM. IgG subclasses are also well known to those in the art and
include, but are not limited to, human IgG1, IgG2, IgG3 and IgG4.
Antibodies can be both naturally occurring and non-naturally
occurring. Furthermore, antibodies include chimeric antibodies,
wholly synthetic antibodies, single chain antibodies, and fragments
thereof. Antibodies may be human or nonhuman. Nonhuman antibodies
may be humanized by recombinant methods to reduce their
immunogenicity in man.
[0098] "Aqueous solution" shall mean any solution in which water is
a solvent. Examples of aqueous solutions include water and
water-based buffer solutions.
[0099] "Complex carbohydrate" shall mean a carbohydrate polymer
comprising more than two types of saccharide monomer units.
Examples of complex carbohydrates include blood group substances
such as Lewis X and Lewis Y.
[0100] "Composition of compounds" at a discrete locus shall mean
the identity of the one or more compounds at that locus. For
example, if locus 1 has compounds A and B, and locus 2 has
compounds A and C, then the composition of compounds at locus 1
differs from that at locus 2.
[0101] "Compound" shall mean any molecule. Compounds include, but
are not limited to, proteins, nucleic acids, glycomers, lipids and
small molecules.
[0102] "Dextran" shall mean a branched polymer of glucose
consisting mainly of .alpha.(1,6)-glycosidic linkages.
[0103] "Discrete locus" shall mean a point, region or area for the
affixation of a compound which does not overlap with another such
point, region or area, and which may further be separated from
another such point, region or area by physical space.
[0104] "Glycomer" shall mean any carbohydrate-containing moiety.
Glycomers include, without limitation, (a) complex carbohydrates,
(b) polysaccharides, (c) oligosaccharides and (d) glycoconjugates.
"Glycoconjugates" include, without limitation, glycoproteins,
glycolipids and glycopolymers. In one embodiment of the glycomer,
the carbohydrate moiety thereof is conjugated, either covalently or
noncovalently, to polyacrylamide.
[0105] "Insoluble protein" shall mean any protein which does not
solubilize in aqueous solution. Examples of insoluble proteins
include trans-membrane proteins.
[0106] "Lectin" shall mean a protein that is capable of
agglutinating erythrocytes, binding sugars, and/or stimulating
mitosis. Examples of lectins include concavalin A.
[0107] "Microarray" shall mean (a) a solid support having one or
more compounds affixed to its surface at discrete loci, or (b) a
plurality of solid supports, each support having one or a plurality
of compounds affixed to its surface at discrete loci. The instant
microarrays can contain all possible permutations of compounds
within the parameters of this invention. For example, the instant
microarray can be an all-glycomer microarray, an all-insoluble
protein microarray, an all-antibody microarray, a disease-specific
microarray, a species-specific microarray, or a tissue-specific
microarray.
[0108] "Nitrocellulose or Hydrogel support" shall mean any solid
support having nitrocellulose or Hydrogel affixed to its surface.
Nitrocellulose or Hydrogel supports include, without limitation,
nitrocellulose-coated or Hydrogel-coated chips (e.g. silicone
chips), slides (e.g. glass slides), filters, plates and beads.
[0109] "Polysaccharide" shall mean a carbohydrate polymer
comprising either one or two types of saccharide monomer units.
Examples of polysaccharides include bacterial cell surface
carbohydrates.
[0110] "Sample", when used in connection with the instant methods,
includes, but is not limited to, any body tissue, skin lesion,
blood, serum, plasma, cerebrospinal fluid, lymphocyte, urine,
exudate, or supernatant from a cell culture.
[0111] "Specifically bind" shall mean the binding of a first entity
to a second entity based on complementarity between the
three-dimensional structures of each. In one embodiment, specific
binding occurs with a K.sub.D of less than 10.sup.-5. In another
embodiment, specific binding occurs with a K.sub.D of less than
10.sup.-8. In a further embodiment, specific binding occurs with a
K.sub.D of less than 10.sup.-11.
[0112] "Subject" shall mean any organism including, without
limitation, a mouse, a rat, a dog, a guinea pig, a ferret, a rabbit
and a primate. In the preferred embodiment, the subject is a human
being.
[0113] Embodiments of the Invention
[0114] This invention provides four microarrays and two articles
useful for making same. The first microarray comprises a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (a) at at least one
discrete locus is affixed a compound selected from the group
consisting of a glycomer, an insoluble protein, a lectin and an
antibody, and (b) the composition of compounds at each discrete
locus differs from the composition of compounds at at least one
other discrete locus.
[0115] The second microarray comprises a plurality of
nitrocellulose or Hydrogel supports, each support having one or a
plurality of compounds affixed to its surface at a single discrete
locus or a plurality of compounds affixed to its surface at
discrete loci, wherein (a) at at least one discrete locus is
affixed a compound selected from the group consisting of a
glycomer, an insoluble protein, a lectin and an antibody, and (b)
the composition of compounds at each discrete locus differs from
the composition of compounds at at least one other discrete
locus.
[0116] The first article comprises a nitrocellulose or Hydrogel
support having dextran affixed to its surface at discrete loci. In
one embodiment the dextran is .alpha.(1,6) dextran.
[0117] The third microarray comprises the first article, wherein at
least one compound is affixed to the dextran at each discrete
locus, the composition of compounds at each discrete locus
differing from the composition of compounds at at least one other
discrete locus.
[0118] The second article comprises a plurality of nitrocellulose
or Hydrogel supports, each support having dextran affixed to its
surface at one or more discrete loci. In one embodiment the dextran
is .alpha.(1,6) dextran.
[0119] The fourth microarray comprises the second article, wherein
at least one compound is affixed to the dextran at each discrete
locus, the composition of compounds at each discrete locus
differing from the composition of compounds at at least one other
discrete locus.
[0120] In one embodiment of the first and third microarrays, the
nitrocellulose or Hydrogel support is selected from the group
consisting of a chip, a slide, a filter, and a plate. In one
embodiment of the second and fourth microarrays, the nitrocellulose
or Hydrogel support is selected from the group consisting of a
chip, a slide, a filter, a plate, and a bead.
[0121] In one embodiment of the above microarrays, the number of
discrete loci is at least 100. In another embodiment, the number of
discrete loci is at least 1000. In a further embodiment, the number
of discrete loci is at least 10,000. In a further embodiment, the
number of discrete loci is at least 50,000.
[0122] In one embodiment of the first and second microarrays, a
glycomer is affixed at at least one locus. In another embodiment,
an insoluble protein is affixed at at least one locus. In another
embodiment, a lectin is affixed at at least one locus. In a further
embodiment, an antibody is affixed at at least one locus. In a
further embodiment, the microarray has affixed to its surface two
or more compounds selected from the group consisting of a glycomer,
an insoluble protein, a lectin and an antibody. In a further
embodiment, the microarray has further affixed to its surface a
compound selected from the group consisting of a soluble protein, a
nucleic acid and a small molecule.
[0123] In one embodiment of the third and fourth microarrays, a
glycomer is affixed to the dextran at at least one locus. In
another embodiment, an insoluble protein is affixed to the dextran
at at least one locus. In another embodiment, a lectin is affixed
to the dextran at at least one locus. In a further embodiment, an
antibody is affixed to the dextran at at least one locus. In
another embodiment, the microarray has affixed to the dextran two
or more compounds selected from the group consisting of a glycomer,
an insoluble protein, a lectin and an antibody. In another
embodiment, the microarray has affixed to its surface a compound
selected from the group consisting of a soluble protein, a nucleic
acid and a small molecule.
[0124] In one embodiment of the instant microarrays, at each locus
is affixed only one compound. In another embodiment, at at least
one locus is affixed a plurality of compounds.
[0125] This invention provides three methods for detecting the
presence of agents in a sample. The first method is a method of
detecting in a sample the presence of one or more agents which
specifically bind to one or more known glycomers, which method
comprises: (a) contacting the sample with the first or second
microarray, wherein each known glycomer is affixed at at least one
discrete locus and wherein the contacting is performed under
conditions which would permit an agent, if present in the sample,
to specifically bind to its corresponding glycomer in the
microarray; and (b) determining whether any known glycomer in the
microarray has an agent specifically bound thereto, thereby
detecting the presence of the one or more agents in the sample.
[0126] The second method is a method of detecting in a sample the
presence of one or more agents which specifically bind to one or
more known insoluble proteins, which method comprises: (a)
contacting the sample with the first or second microarray, wherein
each known insoluble protein is affixed at at least one discrete
locus and wherein the contacting is performed under conditions
which would permit an agent, if present in the sample, to
specifically bind to its corresponding insoluble protein in the
microarray; and (b) determining whether any known insoluble protein
in the microarray has an agent specifically bound thereto, thereby
detecting the presence of the one or more agents in the sample.
[0127] The third method is a method of detecting in a sample the
presence of one or more agents which specifically bind to one or
more known antibodies or lectins, which method comprises: (a)
contacting the sample with the first or second microarray, wherein
each known antibody or lectin is affixed at at least one discrete
locus and wherein the contacting is performed under conditions
which would permit an agent, if present in the sample, to
specifically bind to its corresponding antibody or lectin in the
microarray; and (b) determining whether any known antibody or
lectin in the microarray has an agent specifically bound thereto,
thereby detecting the presence of the one or more agents in the
sample.
[0128] In one embodiment of the above methods, the agent is an
antibody which correlates with a disease. In a further embodiment
of the first method, the agent is an antibody which correlates with
an inflammatory disease. In additional embodiments of the above
methods, the agent is an antibody which correlates with an
infection or the presence of a tumor.
[0129] In one embodiment of the instant methods, the method
comprises detecting the presence of a plurality of agents in the
sample, each of which binds to either a plurality of glycomers, a
plurality of insoluble proteins, or a plurality of lectins or
antibodies, as applicable. In another embodiment of the instant
methods, the method comprises determining the amount of a plurality
of agents in the sample, each of which binds to either one
glycomer, one insoluble protein or one lectin or antibody, as
applicable.
[0130] "Determining" whether an agent is bound to a compound in a
microarray can be performed according to methods well known in the
art. Such methods include, but are not limited to, fluorescence,
radioimmunoassay, and immunolabeling detection.
[0131] In the instant methods of detection, several embodiments are
provided which include, without limitation, the following: (a) one
agent in a sample binds to one compound on the instant microarray;
(b) one agent in a sample is detected that binds to more than one
compound on the microarray; (c) the collective presence of a
plurality of agents in a sample is detected, wherein each such
agent binds to one or more compounds on the microarray; and (d)
each of a plurality of agents in a sample is individually detected,
wherein each such agent binds to one or more compounds on the
microarray.
[0132] This invention further provides three quantitative methods.
The first method is a method of determining the amount of one or
more agents in a sample, each of which specifically binds to one or
more known glycomers, which method comprises: (a) contacting the
sample with the first or second microarray, wherein each known
glycomer is affixed at at least one discrete locus, and wherein the
contacting is performed under conditions which would permit an
agent, if present in the sample, to specifically bind to its
corresponding glycomer in the microarray; (b) for each known
glycomer in the microarray, determining the amount of agent
specifically bound thereto; and (c) comparing the amounts so
determined to a known standard, thereby determining the amount of
the one or more agents in the sample.
[0133] The second method is a method of determining the amount of
one or more agents in a sample, each of which specifically binds to
one or more known insoluble proteins, which method comprises: (a)
contacting the sample with the first or second microarray, wherein
each known insoluble protein is affixed at at least one discrete
locus, and wherein the contacting is performed under conditions
which would permit an agent, if present in the sample, to
specifically bind to its corresponding insoluble protein in the
microarray; (b) for each known insoluble protein in the microarray,
determining the amount of agent specifically bound thereto; and (c)
comparing the amounts so determined to a known standard, thereby
determining the amount of the one or more agents in the sample.
[0134] The third method is a method of determining the amount of
one or more agents in a sample, each of which specifically binds to
one or more known antibodies or lectins, which method comprises:
(a) contacting the sample with the first or second microarray,
wherein each known antibody or lectin is affixed at at least one
discrete locus, and wherein the contacting is performed under
conditions which would permit an agent, if present in the sample,
to specifically bind to its corresponding antibody or lectin in the
microarray; (b) for each known antibody or lectin in the
microarray, determining the amount of agent specifically bound
thereto; and (c) comparing the amounts so determined to a known
standard, thereby determining the amount of the one or more agents
in the sample.
[0135] In one embodiment of the instant quantitative methods, the
agent is an antibody which correlates with a disease. In a further
embodiment of the first method, the agent is an antibody which
correlates with an inflammatory disease. In additional embodiments
of the above methods, the agent is an antibody which correlates
with an infection or the presence of a tumor.
[0136] In one embodiment of the instant quantitative methods, the
method comprises determining the amount of a plurality of agents in
the sample, each of which binds to either a plurality of glycomers,
a plurality of insoluble proteins, or a plurality of lectins or
antibodies, as applicable. In another embodiment of the instant
quantitative methods, the method comprises determining the amount
of a plurality of agents in the sample, each of which binds to
either one glycomer, one insoluble protein or one lectin or
antibody, as applicable.
[0137] "Determining" the amount of an agent which is bound to a
compound in a microarray can be performed according to well known
methods in the art. The "known standards" useful for the instant
quantitative methods include, for example, correlations between
known concentrations of agents in a control sample and their
corresponding values as determined using the instant
microarray.
[0138] In the instant quantitative methods, several embodiments are
provided which include, without limitation, the following: (a) one
agent in a sample binds to one compound on the instant microarray;
(b) one agent in a sample is quantitated that binds to more than
one compound on the microarray; (c) the collective amount of a
plurality of agents in a sample are quantitated, wherein each such
agent binds to one or more compounds on the microarray; and (d)
each of a plurality of agents in a sample is individually
quantitated, wherein such agent binds to one or more compounds on
the microarray.
[0139] This invention further provides three diagnostic methods.
The first method is a method of determining whether a subject is
afflicted with a disorder characterized by the presence or absence
in an afflicted subject of an agent which specifically binds to a
known glycomer, which method comprises: (a) contacting a suitable
sample from the subject with the first or second microarray,
wherein the known glycomer is affixed at at least one discrete
locus and wherein the contacting is performed under conditions
which would permit the agent, if present in the sample, to
specifically bind to the known glycomer in the microarray; and (b)
determining whether the known glycomer in the microarray has the
agent specifically bound thereto, thereby determining whether the
subject is afflicted with the disorder.
[0140] The second method is a method of determining whether a
subject is afflicted with a disorder characterized by the presence
or absence in an afflicted subject of an agent which specifically
binds to a known insoluble protein, which method comprises: (a)
contacting a suitable sample from the subject with the first or
second microarray, wherein the known insoluble protein is affixed
at at least one discrete locus and wherein the contacting is
performed under conditions which would permit the agent, if present
in the sample, to specifically bind to the known insoluble protein
in the microarray; and (b) determining whether the known insoluble
protein in the microarray has the agent specifically bound thereto,
thereby determining whether the subject is afflicted with the
disorder.
[0141] The third method is a method of determining whether a
subject is afflicted with a disorder characterized by the presence
or absence in an afflicted subject of an agent which specifically
binds to a known antibody or lectin, which method comprises: (a)
contacting a suitable sample from the subject with the first or
second microarray, wherein the known antibody or lectin is affixed
at at least one discrete locus and wherein the contacting is
performed under conditions which would permit the agent, if present
in the sample, to specifically bind to the known antibody or lectin
in the microarray; and (b) determining whether the known antibody
or lectin in the microarray has the agent specifically bound
thereto, thereby determining whether the subject is afflicted with
the disorder.
[0142] In one embodiment of the instant diagnostic methods, the
subject is human. In one embodiment of the first method, the
disorder is an inflammatory disorder. In another embodiment of the
first method, the inflammatory disorder is celiac disease. In one
embodiment of the third method, the disorder is HIV-1
infection.
[0143] The following are specific examples of the instant
diagnostic methods. In the first example, a subject's serum is
analyzed for the presence of HIV-1 gp120 and IgG-anti-HIV-1 gp120,
the presence of both indicating active HIV-1 infection. In the
second example, a subject's serum is analyzed for the presence of
either HIV-1 gp120 and IgG-anti-HIV-1 gp120, the absence of the
HIV-1 gp120 and the presence of IgG-anti-HTV-1 gp120 antibody
indicating HIV-1 infection or immunization. In the third example, a
subject's serum is analyzed for the presence of HIV-1 gp120 and
IgG-anti-HIV-1 gp120, the absence of both indicating that the
subject is neither HIV-1 infected nor immunized. In the fourth
example, a subject's serum is analyzed for the presence of
IgA-anti-gliadin and IgA-anti-TGt, the presence of both indicating
that the subject is afflicted with celiac disease. Finally, in the
fifth example, a subject's serum is analyzed for the presence of
IgA-anti-gliadin, the presence of this antibody indicating the
possibility that the subject is afflicted with celiac disease.
[0144] This invention further provides a method of determining
whether an antibody known to specifically bind to a first glycomer
also specifically binds to a second glycomer, which method
comprises: (a) contacting the antibody with the first or second
microarray, wherein a plurality of glycomers, other than the first
glycomer, are affixed at discrete loci in the microarray, and
wherein the contacting is performed under conditions which would
permit the antibody to specifically bind to the first glycomer if
it were present in the microarray; and (b) determining, whether any
of the glycomers in the microarray, other than the first glycomer,
has the antibody specifically bound thereto, thereby determining
whether the antibody also specifically binds to a second
glycomer.
[0145] This invention further provides a method of determining
whether an antibody known to specifically bind to a first insoluble
protein also specifically binds to a second insoluble protein,
which method comprises: (a) contacting the antibody with the first
or second microarray, wherein a plurality of insoluble proteins,
other than the first insoluble protein, are affixed at discrete
loci in the microarray, and wherein the contacting is performed
under conditions which would permit the antibody to specifically
bind to the first insoluble protein if it were present in the
microarray; and (b) determining whether any of the insoluble
proteins in the microarray, other than the first insoluble protein,
has the antibody specifically bound thereto, thereby determining
whether the antibody also specifically binds to a second insoluble
protein.
[0146] This invention further provides a method of making a
microarray comprising a nitrocellulose or Hydrogel support having
affixed to its surface at discrete loci a plurality of compounds,
which method comprises contacting the nitrocellulose or Hydrogel
support with the compounds under suitable conditions, whereby (a)
at at least one discrete locus is affixed a compound selected from
the group consisting of a glycomer, an insoluble protein, a lectin
and an antibody, and (b) the composition of compounds at each
discrete locus differs from the composition of compounds at at
least one other discrete locus.
[0147] This invention further provides a method of making a
microarray comprising a plurality of nitrocellulose or Hydrogel
supports, each support having one or a plurality of compounds
affixed to its surface at a single discrete locus or a plurality of
compounds affixed to its surface at discrete loci, which method
comprises contacting the nitrocellulose or Hydrogel supports with
the compounds under suitable conditions, whereby (a) at at least
one discrete locus is affixed a compound selected from the group
consisting of a glycomer, an insoluble protein, a lectin and an
antibody, and (b) the composition of compounds at each discrete
locus differs from the composition of compounds at at least one
other discrete locus.
[0148] This invention further provides a method of making the first
article comprising contacting a nitrocellulose or Hydrogel support
with dextran at discrete loci under suitable conditions.
[0149] In one embodiment of this method, the method further
comprises the step of affixing at least one compound to the dextran
at each discrete locus, whereby the composition of compounds at
each discrete locus differs from the composition of compounds at at
least one other discrete locus.
[0150] This invention further provides a method of making the
second article comprising contacting a plurality of nitrocellulose
or Hydrogel supports with dextran, whereby each support has dextran
affixed to its surface at one or more discrete loci.
[0151] In one embodiment of this method, the method further
comprises the step of affixing at least one compound to the dextran
at each discrete locus, whereby the composition of compounds at
each discrete locus differs from the composition of compounds at at
least one other discrete locus.
[0152] This invention further provides six kits. The first kit
comprises one of the instant microarrays and instructions for use.
The second kit comprises one of the instant microarrays and a
desiccant. The third kit comprises one of the instant microarrays
immersed in an aqueous solution.
[0153] The fourth kit is a kit for practicing the first diagnostic
method, which comprises: (a) a microarray comprising a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (i) at at least one
discrete locus is affixed the glycomer to which the agent present
or absent in an afflicted subject specifically binds, and (ii) the
composition of compounds at each discrete locus differs from the
composition of compounds at at least one other discrete locus; and
(b) instructions for use.
[0154] The fifth kit is a kit for practicing the second diagnostic
method, which comprises: (a) a microarray comprising a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (i) at at least one
discrete locus is affixed the insoluble protein to which the agent
present or absent in an afflicted subject specifically binds, and
(ii) the composition of compounds at each discrete locus differs
from the composition of compounds at at least one other discrete
locus; and (b) instructions for use.
[0155] The sixth kit is a kit for practicing the third diagnostic
method, which comprises: (a) a microarray comprising a
nitrocellulose or Hydrogel support having affixed to its surface at
discrete loci a plurality of compounds, wherein (i) at at least one
discrete locus is affixed the antibody or lectin to which the agent
present or absent in an afflicted subject specifically binds, and
(ii) the composition of compounds at each discrete locus differs
from the composition of compounds at at least one other discrete
locus; and (b) instructions for use.
[0156] This invention further provides a first antibody capable of
specifically binding to a glycomer present on the surface of a
mammalian macrophage, which glycomer, or structural mimic thereof,
is also endogenous to, and present on the surface of, a bacterial
cell. In one embodiment, the antibody is a groove-type antibody. In
another embodiment, the antibody is designated 4.3.F1 (ATCC
Accession No. PTA-3259). In a further embodiment, the antibody is
designated 45.21.1 (ATCC Accession No. PTA-3260).
[0157] This invention further provides a second antibody capable of
specifically binding to a glycomer present on the surface of a
mammalian intestinal epithelial cell, which glycomer, or structural
mimic thereof, is also endogenous to, and present on the surface
of, a bacterial cell. In one embodiment, the antibody is a
cavity-type antibody. In another embodiment, the antibody is
designated 16.4.12E (ATCC Accession No. PTA-3261).
[0158] This invention further provides a method of determining
whether a subject is afflicted with a disorder characterized by the
presence of a glycomer on the surface of macrophages in an
afflicted subject, which glycomer, or structural mimic thereof, is
also endogenous to, and present on the surface of, a bacterial
cell, comprising: (a) contacting a sample of the subject's
macrophages with the first antibody; and (b) determining whether
the antibody specifically binds to the macrophages in the sample,
such binding indicating that the subject is afflicted with the
disorder. In one embodiment the subject is human. In another
embodiment, the disorder is an immune disorder or an inflammatory
disorder.
[0159] Finally, this invention provides a method of determining
whether a subject is afflicted with a disorder characterized by the
presence of a glycomer on the surface of intestinal epithelial
cells in an afflicted subject, which glycomer, or structural mimic
thereof, is also endogenous to, and present on the surface of, a
bacterial cell, comprising: (a) contacting a sample of the
subject's intestinal epithelial cells with the second antibody; and
(b) determining whether the antibody specifically binds to the
intestinal epithelial cells in the sample, such binding indicating
that the subject is afflicted with the disorder. In one embodiment
the subject is human. In another embodiment, the disorder is an
immune disorder or an inflammatory disorder. In a further
embodiment, the disorder is celiac disease. In one embodiment of
each facet of this invention (i.e., each of the instant
compositions of matter and methods), the glycomer, insoluble
protein, lectin or antibody of the instant microarray is conjugated
with a green fluorescent protein (GFP), and this conjugate is
affixed to a nitrocellulose or Hydrogel support either via the GFP
moiety or not. "Conjugation" as used herein includes, without
limitation, conjugation by covalent or noncovalent means. In a
further embodiment, conjugation of a second protein with GFP is
accomplished via formation of a fusion protein with the second
protein. As used herein, "a fusion protein" includes, for example,
a protein encoded by a gene that comprises a GFP-encoding region
and a non-GFP-encoding region. As used herein, GFP includes,
without limitation, whole GFP as well as fluorescent fragments
thereof. In a further embodiment of each facet of this invention,
GFP alone (i.e., unconjugated form), is bound to the instant
microarray.
[0160] This invention will be better understood from the
Experimental Details that follow. However, one skilled in the art
will readily appreciate that the specific methods and results
discussed are merely illustrative of the invention as described
more fully in the claims which follow thereafter.
EXPERIMENTAL DETAILS
[0161] First Series of Experiments
[0162] This invention provides novel antigen- and antibody-based
microarrays for monitoring and quantifying a broad spectrum of
biological molecules and their molecular interactions. A microarray
technique is used to spot thousands of antigens and/or antibodies
on a solid surface. This strategy can be applied to any molecular
target, including naturally occurring proteins, carbohydrates,
lipids and nucleic acids, as well as synthetic compounds. The
instant microarray is useful for monitoring the expression of
specific antibodies and other cellular factors in body fluids, and
is therefore useful for disease diagnosis and basic immunological
investigation. When a large repertoire of distinct monoclonal
antibodies are arrayed, an antibody library microarray is produced.
Application of these microarrays for global analysis of gene
expression at the translational and post-translational levels is
envisioned. The Elvin A. Kabat Collection of antigens and
antibodies at Columbia University is useful in practicing this
technology.
[0163] I. Materials, Methods and Results
[0164] (A) Method for Antigen/Antibody Immobilization
[0165] It is experimentally demonstrated here that
nitrocellulose-coating can serve as a suitable matrix for
immobilizing polysaccharides, glycoprotein, glycolipid, protein and
antibodies on a glass surface without chemical conjugation. A set
of commercially available glass slides, including those coated with
nitrocellulose (ONCYTE Film-Slides, Grace Bio-Labs, Inc., Bend,
Oreg.), poly-L-lysine (POLY-PREP.TM., Sigma), aminoalkylsilane
(SILANE-PREP.TM., Sigma) and regular micro-glass slides, were
compared for their capacity to immobilize macromolecules of
distinct structural properties. In the initial experiments,
fluorescence-conjugated dextran molecules were applied. An extended
panel of antigen preparations, including polysaccharide,
glycoprotein, glycolipid, protein and antibody, were then
investigated. Examples of these investigations are shown in FIGS. 1
and 2.
[0166] (B) Printing and Long-Term Storage of Antigen/Antibody
Microarrays
[0167] A high precision robot designed to produce cDNA microarrays
(GMS 417 Arrayer, Genetic Microsystems, Inc., Woburn Mass.) was
used to spot carbohydrate antigens onto a glass slide pre-coated
with nitrocellulose polymer (ONCYTE Film-Slides, Grace Bio-Labs,
Inc., Bend, Oreg.). Spots of antigens were printed with spot sizes
at approximately 200 micron and 400 micron intervals,
center-to-center. They were air dried and stored at room
temperature before use. Conditions for long-term storage of the
printed antigen/antibody microarrays were compared. The results
show that (1) the air-dried carbohydrate microarrays can be stored
at room temperature for at least one year without significant
inactivation of their immunological activities; and (2) the
antibody microarrays can be stored in an aqueous solution at
4.degree. C. for at least one year without significant decrease in
their antigen-binding activity, as illustrated using antibodies
with anti-carbohydrate specificity.
[0168] (C) Proper Macromolecules for Coupling Smaller Bioreactive
Molecules and Displaying them on a Solvent-Accessible Surface
[0169] Dextran preparations, especially .alpha.(1,6)dextrans, can
serve as carrier molecules to conjugate other biologically active
molecules. Such dextran-containing conjugates can be immobilized on
a nitrocellulose-surface without further chemical conjugation. This
method was demonstrated by immobilizing the
fluorescence-.alpha.(1,6)dextran conjugates of different molecular
weights, ranging from 35 kD to 2000 kD, on a nitrocellulose-coated
glass slide. The fluorescence group is accessible to the
anti-fluorescence antibodies in solution, thereby demonstrating
specific binding. Methods for producing dextran-conjugates for
surface immobilization follow.
[0170] (1) Mild Sodium Periodate Oxidization of Dextran to Create
Highly Reactive Aldehyde Functional Groups (CHO)
[0171] .alpha.(1,6)dextran, preparation N279 (B512), was dissolved
in 0.01M NaAcetate buffer, pH 5.5, at 10 mg/ml and warmed in a
37.degree. C. water bath for 30 minutes. NaIO.sub.4 was then added
to the final concentration of 1.times.10.sup.-2 M. The solution was
mixed well and left to stand at room temperature for one hour in
the dark. The preparation was dialyzed against 0.02M BBS (Borate
buffered saline) pH 8.0, 4.degree. C., overnight.
[0172] (2) Mild Oxidization of Carbohydrate Structure of IgG to
Create CHO Group for Surface Immobilization
[0173] The above protocol was also applied to produce active CHO
groups in carbohydrate molecules that exist naturally in the C
region of IgG molecules. Such CHO-activated IgG is then covalently
linked to the amino group of an amino-dextran molecule that was
immobilized on a nitrocellulose-coated glass slide. Preparations of
amino-dextrans are commercially available (Molecular Probes,
Eugene). This method of IgG immobilization uses a carbohydrate
structure attached in the Fc region of IgG, which is away from the
antibody combining sites of antibody molecule, and thus preserves
the antibodies' binding activity.
[0174] (3) Coupling of Biotin-LC-Hydrazide to the Oxidized
Dextran
[0175] The above oxidized dextran was diluted 10-fold in 0.1M
NaAcetate, pH 5.5. A 1/3 volume of 5 mM Biotin-LC-Hydrazide was
added dropwise. The mixture was shaken at room temperature for one
hour. The reaction was terminated by addition of 0.5 ml of 1M Tris
HCl, pH 7.5, and the mixture was then dialyzed against Tris buffer
(0.1M Tris pH 7.5, 0.1 M NaCl, 2.0 mM MgCl2). The biotinylated
dextrans are then ready to be immobilized on nitrocellulose-coated
glass slides so that their biotin-groups are accessible to other
molecules in solution.
[0176] (4) Surface Immobilization of Biotinylated Molecules
[0177] Standard methods were applied to couple NHS-Biotin (BRL
#5533LA) to target molecules, i.e., either protein or
polysaccharide. The biotinylated molecule was incubated with avidin
at a proper molar ratio depending on the molecular weight of the
target and its molar ratio with biotin. The molecules were then
spotted on a surface that was precoated with biotin-dextran and
blocked with BSA or gelatin. This strategy allows a flexible
arrangement of antigen or antibody microarrays for a desired
purpose, and avoids non-specific binding of target molecules on a
surface.
[0178] (5) Gluttaraldehyde-Conjugation to Generate Dextran-Small
Molecule Conjugates
[0179] Small bioactive molecules containing amine group(s) can be
coupled to amino-dextrans (Molecular Probes) by glutaraldehyde.
Glutaraldehyde was added to the mixtures of the target molecule and
amino-dextran (at proper molar ratios) to a final concentration of
0.2%. They were at room temperature for two hours. The reaction was
stopped by addition of 1M ethanolamine at 6.1 .mu.l/ml. The mixture
was at room temperature for an additional two hours and then
dialyzed against 1.times. PBS or other proper solution, overnight.
The dextran-small molecule conjugates were then spotted on
nitrocellulose-coated slides. This method is suitable for
generating microarrays having a large repertoire of small molecules
and useful for high throughput drug screening or other biomedical
investigations.
[0180] (D) Solutions for Preparing Macromolecules of Distinct
Physicochemical Properties
[0181] (1) Solution Storage
[0182] Microbial polysaccharides, blood group substances and
glycolipids were solubilized and stored in saline at a
concentration of about 1 mg/ml at 4.degree. C. A small droplet of
chloroform was added to prevent microbes. In this simple way, most
solutions can be stored for years. The agents were diluted in
saline at required concentrations immediately before spotting.
[0183] (2) Protein Antigens
[0184] Soluble protein preparations were prepared at relatively
high concentrations in 1.times. PBS (mg/ml) with addition of 20%
glycerol and frozen at -80.degree. C. They were diluted in 1.times.
PBS before use and stored at 4.degree. C. for a short period of
time (a few days). Antibody preparations were generally stored at
4.degree. C. in 1.times. PBS except in special cases. Some E.
Coli-expressed protein antigens are water-insoluble. The
preparations were purified in a denatured condition and stored them
in the same solution at 4.degree. C. The freezing process is
avoided for these proteins. In most cases these preparations can be
immobilized on a nitrocellulose matrix without special treatment.
This method has been applied successfully in applicant's laboratory
for hybridoma screening.
[0185] (E) Staining and Scanning of Antigen/Antibody
Microarrays
[0186] (1) General Application Protocol
[0187] Immediately before use, the printed antigen/antibody
microarrays were rinsed with 1.times. PBS with 0.05% Tween 20 and
then blocked by incubating the slides in 1% BSA in PBS containing
0.05% NaN.sub.3 at 37.degree. C. for 30 minutes. The microarrays
were then incubated at room temperature with fluorescent-antibody
conjugate at proper titration in 1% BSA PBS containing 0.05%
NaN.sub.3 and 0.05% Tween 20. Slides were rinsed with 1.times. PBS
with 0.05% Tween 20 five times, air-dried at room temperature and
then scanned for fluorescent signals. A ScanArray 5000 Standard
Biochip Scanning System (GSI Lumonics, Inc. and Packard BioChip
Technologies, Inc.) which is equipped with multiple lasers,
emission filters, ScanArray Acquisition Software and QuantArray
Microarray Analysis Software, was used to scan the stained antigen
microarray, quantify spot-associated fluorescent-signals and
analyze data.
[0188] (2) Special Application Protocols
[0189] (a) Method for Improving Signal Detection
[0190] A technical problem that has limited the application of the
nitrocellulose-coated glass slide is its association with "white
color" and non-specific fluorescent signals upon scanning. This
problem is solved using the method that follows. (a) After staining
a microarray glass slide, allow it to air-dry for a few minutes (at
this stage, the nitrocellulose-coated region is white in color).
(b) Soak the slide in 100% ethanol for 1-2 minutes until the white
color of the nitrocellulose-coated region disappears and the entire
slide becomes transparent. (c) Quickly spin the slide to remove
extra ethanol. (d) Scan the slide when it is completely
transparent. After storage for a few days or longer, depending on
the humidity level in the air, the slide may turn back to the white
color. One may repeat the above process to make the slide
transparent again if necessary.
[0191] (b) Staining Antigen/Antibody Microarrays Using
Non-Fluorescent Dyes
[0192] The antigen/antibody microarray can be stained with
antibodies, antigens or other reactors that are conjugated with
non-fluorescent dyes. The commonly used alkaline phosphatase (AP)
and peroxidase are useful alternatives. One may practice this
method as follows. (a) Stain an antigen microarray with a human
serum sample at the proper dilution as described above. (b) After
washing, stain the slide with an AP-conjugated anti-human IgG
antibody at the proper dilution. (c) Wash the slide and develop the
color by adding AP substrate BCTP alone or BCIP plus NBT. (d) Stop
the reaction by adding a Tris-EDTA solution (20 mM Tris, pH 7.5, 5
mM EDTA). (e) Rinse the slide with distilled water and then scan it
with a non-flourescent slide scanner, or observe the color reaction
under a regular microscope.
[0193] (F) Sensitivity of Antigen Microarrays
[0194] Sensitivity is one of the critical parameters that determine
the diagnostic value of antigen microarrays. In preliminary
experiments (FIG. 1), dextran preparations were arrayed on
nitrocellulose-coated glass slides in concentrations ranging from
100 .mu.g/ml to 3.3 ng/ml. They were stained with
fluorescence-labeled anti-dextran mAbs, either 4.3.F1 or 16.4.12E,
at a concentration of 1 .mu.g/ml. In both cases, the captured
fluorescent signals (intensities) detected were positively
correlated to the antigen concentrations. For example, 4.3.F1 was
only detectable when the concentrations of N279, LD7 and B1299S are
higher than 0.4, 20 or 100 .mu.g/ml, respectively (spots of
antigens with concentrations lower than 0.4 .mu.g/ml were not shown
in FIG. 1). Signal saturation was not observed even at the highest
concentration of 100 .mu.g/ml, showing the technical potential for
improving the sensitivity of antigen microarrays.
[0195] (G) Epitope-Specific Antigen Microarray
[0196] Glycoconjugate technology was used to produce an
"epitope-specific antigen microarray" A naturally occurring antigen
may be composed of multiple antigenic determinants. Frequently, one
or a few of them serve as predominant antigenic determinants for
host recognition. Antibodies specific for some carbohydrate
epitopes of a microbial polysaccharide can be more protective than
those reacting with others. There are also cross-reactive antigenic
determinants that may be shared among strains or even species of
microorganisms. Such cross-reactivities may cause difficulty in
typing corresponding infectious agents. Thus, it is useful to
define the fine specificities of antibodies elicited by an
infection or by vaccination.
[0197] To produce an epitope-specific microarray, preparations of
glycoproteins displaying .alpha.(1,6)-linked glucoses, i.e.,
isomaltotriose-coupled BSA (IM3-BSA) or its KLH-conjugate
(IM3-KLH), were applied in a microarray experiment. These
conjugates have the terminal non-reducing end epitope of
.alpha.(1,6)dextran in common but differ in their protein carriers.
As expected, mAb 16.4.12E (cavity-type), but not mAb 4.3.F1
(groove-type), bind to the microspots of IM3-protein conjugates
(FIGS. 1E and 2E). Thus, glycoproteins can be immobilized on a
nitrocellulose-coated glass slide and their antigenic determinants
remain accessible to antibodies in solution.
[0198] Neoglycolipids, i.e., glycolipid conjugates, were used to
produce epitope-specific microarrays, wherein
stearylamine-isomaltosyl oligosaccharide conjugates, ST-IM3, ST-IM5
and ST-IM7, were applied (data not shown). Such glycolipid
conjugates are homogeneic, since each lipid molecule can only be
coupled by a single oligosaccharide. Unlike a glycoprotein
conjugate, the sugar chain can be conjugated on to multiple sites
of a protein molecule, generating a heterogenic population of
antigenic determinants. Both sugar epitope and the amino acid
residues adjacent to it may be structurally involved in forming
these antigenic determinants.
[0199] (H) Antibody Microarrays
[0200] Experiments were performed to develop antibody-based
microarrays. A panel of anti-dextran mAbs were immobilized at a
concentration of 0.5 mg/ml on a set of glass slides. These included
(a) a nitrocellulose-coated glass slide (ONCYTE Film-Slides, Grace
Bio-Labs, Inc., Bend, Oreg.); (b) a poly-lysine treated slide
(POLY-PREP.TM., Sigma); (c) a silane-treated glass slide
(SILANE-PREP.TM., Sigma); and (d) an un-treated, pre-cleaned glass
slide. These slides were then reacted with fluorescence-tagged
dextran preparations of distinct structures. Only the
nitrocellulose-slides showed spots of specific fluorescent signals.
Thus, anti-dextran mAbs arrayed on the nitrocellulose-glass slide
retained their antigen binding specificities. As described above,
the antibody microarrays can be stored at room temperature in an
air-dried condition for a few months and maintain their
antibody-binding activity. The antibodies investigated include
monoclonal antibodies of different isotypes (IgM, IgG and IgA).
[0201] II. Application of Antigen-Antibody Biochip Technology for
Basic Research and Clinical Investigation
[0202] (A) Mapping Antibody Combining Sites by Applying
Antigen/Antibody Microarrays
[0203] A well-established antigen-antibody system, i.e., dextrans
and anti-dextran monoclonal antibodies (mabs) [10, 11], was used
for these investigations. A panel of purified dextran preparations
of different linkage compositions and of different ratios of
terminal/internal epitopes [10] were immobilized on
nitrocellulose-coated glass slides and then incubated with
monoclonal antibodies of defined specificities, either the
cavity-type or the groove-type anti-dextrans [12]. The former is
specific for the terminal non-reducing end structure of
.alpha.(1,6)dextran; the latter recognizes the internal linear
chain of the polysaccharide. When a cavity-type mAb, 16.4.12E, was
applied on the glass slide, it bound to the immobilized
.alpha.(1,6)dextran preparations having branches, but not those
with only internal linear chain structures. By contrast, a
groove-type mAb, 4.3.F1, bound to the dextran preparations
dominated by linear chain structures, but not to the heavily
branched .alpha.(1,6)dextrans (FIG. 1). Thus, the use of
polysaccharides of defined structure in microarrays is an important
strategy for studying antibody specificities and for characterizing
immune responses to microbial infections.
[0204] As described above, methods to produce epitope-specific
microarrays were also established, allowing characterization of the
fine specificity of antibodies. Considering the presence of both
the internal chain epitopes and the terminal non-reducing end
epitopes in many polysaccharides, application of both
polysaccharide macromolecules and their oligosaccharide conjugates
on the instant microarrays will significantly enhance the power of
the system, including its sensitivity, specificity and the
detecting repertoires of antigenic determinants.
[0205] (B) Investigation of the Specificity and Cross-Reactivity of
Antibodies/Receptors
[0206] An advantage of the instant microarray is that it provides a
high throughput strategy to characterize the specificity and
cross-reactivity of an antibody or a lectin molecule. FIG. 2 shows
an example of this approach. A collection of about 50
carbohydrate-containing antigens, including microbial
polysaccharides, blood group substances and other glyco-conjugates,
were arrayed on the nitrocellulose-coated slide. They were then
stained with anti-dextran mabs, 4.3.F1, or 16.4.12E. As shown in
FIG. 2, certain cross-reactive signals were detected for a few
antigen preparations, being spots 2a, 3a and 6a for mAb 4.3.F1, and
2a, 3a, 4a, 6a, 1e, and 2e for 16.4.12E. Antigens arrayed at these
locations are: 2a. Klebsiella polysaccharide type 11; 3a.
Klebsiella type 13; 4a. Klebsiella type 21; 6a. Chondroitin sulfate
B polysaccharide; 1e. IM3-BSA and 2e. IM3-KLH.
[0207] As discussed above, the binding of 16.4.12E (cavity-type),
but not 4.3.F1 (groove-type), to IM3-BSA (1e) and IM3-KLH (2e)
reflects the epitope-specific binding activities of these two
monoclonal antibodies. Their binding to other antigens is, however,
unexpected. The fluorescent intensities of cross-reactivities are
much weaker than the binding to .alpha.(1,6)dextran N279. They were
detected at a high antigen concentration (500 .mu.g/ml)
corresponding to or lower than the signals of specific bindings at
much lower antigen concentrations (0.8 to 4 .mu.g/ml). On an ELISA
plate, such weak cross-reactivities are not detectable (data not
shown). The cross-reactivities to CS-B polysaccharide (FIG. 2A and
FIG. 2B), is interesting since CS-B is not a microbial antigen and
was prepared from the intestinal mucosa of porcine. Further
investigation of such reactivity led to the discovery of Dex-IdX as
a cell type-specific marker in mouse and human (FIGS. 3-5).
[0208] (C) Clinical Application of Antigen Microarrays
[0209] The antigen-based microarray can be applied for detection
and, characterization of a wide range of microbial infections.
During infection, whether viral, bacterial, fungal or parasitic,
the host usually responds with the formation of antibodies which
can be detected by a modified version of any of the methods used
for antigen detection. The formation of antibodies and their time
course depends on the antigenic stimulation provided by the
infection. Recognition of these patterns provides evidence of
recent or past infection. Microarrays of a large panel of antigens
allow detection of many specificities in a single assay and thus
allow a rapid diagnosis of infections. The diagnostic power of the
instant microarrays will only increase as more microbial antigens
and/or their antigenic determinants are characterized and
applied.
[0210] To demonstrate this principle, a small scale antigen
microarray with about 50 antigens was produced (FIG. 2) and tested
with human sera from normal individuals and celiac patients. These
specimens were diluted at 1:20 in 1% BSA-PBS with 0.025% Tween 20
and applied to the microarray. The bound human antibodies were
visualized by application of a second antibody specific for human
IgG or IgA that was conjugated with a fluorescent molecule (an
anti-human IgG.sup.Cy3 and an anti-human IgA.sup.Cy5). The positive
staining of the microspots of these arrayed antigens by the serum
specimens ranged from 6% (three of fifty spots were detected) to
12% (six of fifty spots were detected). The antigens detected were
mainly microbial polysaccharides, including Klebsiella
polysaccharide type 7, 13, 14, 21 and 33, Dudmans Rhizobium
Trifelli TA1 and Levan.
[0211] The majority of the positive staining was of human IgG
antibodies, although IgAs were also detected. This investigation
has demonstrated that the instant microarray has the sensitivity to
detect specific antibodies in human serum. Given the current
capacity of microspotting technology in the art, about twenty
thousand antigens can be arrayed on a single glass slide. It is
expected, therefore, that this microarray is capable of
characterizing a wide range of microbial infections using very
limited samples in a single experiment.
[0212] III. Discussion
[0213] The instant microarrays are significantly different from
known cDNA microarray and oligo-chip technologies, which target
only nucleic acids. Applicant has employed high throughput
microarray technology to develop a novel strategy for detecting,
quantifying and characterizing proteins, carbohydrates and other
biological molecules, and useful in the new areas of post-genomic
research, namely proteomics and glycomics.
[0214] Differing from current immunoassays which detect specific
molecules one-by-one, this technology is designed to detect and
quantify a large repertoire of distinct biological molecules in a
single assay. Combining a high throughput microarray technique and
a sensitive confocal fluorescent scanning method, this technology
is useful for detecting thousands of distinct molecules using a
small amount of biological specimen, such as a drop of blood or
other bodily fluid. This technology can be extended to the
genome-wide scanning of protein expression and post-translational
modification.
[0215] References
[0216] 1. Ramsay, G., DNA chips: state-of-the art. Nature
Biotechnology, 1998. 16(1): p. 40-4.
[0217] 2. Brown, P. O. and D. Botstein, Exploring the new world of
the genome with DNA microarrays. Nature Genetics, 1999. 21(1
Suppl): p. 33-7.
[0218] 3. DeRisi, J. L., V. R. Iyer, and P. O. Brown, Exploring the
metabolic and genetic control of gene expression on a genomic
scale. Science, 1997. 278(5338): p. 680-6.
[0219] 4. Lueking, A., et al., Protein microarrays for gene
expression and antibody screening. Analytical Biochemistry, 1999.
270(1): p. 103-11.
[0220] 5. Ge, H., UPA, a universal protein array system for
quantitative detection of protein-protein, protein-DNA, protein-RNA
and protein-ligand interactions. Nucleic Acids Research, 2000.
28(2): p. e3.
[0221] 6. Varki, A., Biological roles of oligosaccharides: all of
the theories are correct. Glycobiology, 1993. 3(2): p. 97-130.
[0222] 7. Feizi, T., Carbohydrate recognition systems in innate
immunity. Advances in Experimental Medicine & Biology, 1998.
435: p. 51-4.
[0223] 8. Feizi, T. and R. W. Loveless, Carbohydrate recognition by
Mycoplasma pneumoniae and pathologic consequences. American Journal
of Respiratory & Critical Care Medicine, 1996. 154(4 Pt 2): p.
S133-6.
[0224] 9. Zareba, T. W., et al., Binding of extracellular matrix
proteins by enterococci. Curr Microbiol, 1997. 34(1): p. 6-11.
[0225] 10. Wang, D. and E. A. Kabat, Carbohydrate Antigens
(Polysaccharides), in Structure of Antigens, M. H. V. V.
Regenmortal, Editor. 1996, CRC Press: Boca Raton New York London
Tokyo. p. 247-276.
[0226] 11. Wang, D., et al., The repertoire of antibodies to a
single antigenic determinant. Molecular Immunology, 1991. 28(12):
p. 1387-97.
[0227] 12. Cisar, J., et al., Binding properties of immunoglobulin
combining sites specific for terminal or nonterminal antigenic
determinants in dextran. J. Exp. Med., 1975. 142: p. 435-459.
[0228] 13. Matsuda, T. and E. A. Kabat, Variable region cDNA
sequences and antigen binding specificity of mouse monoclonal
antibodies to isomaltosyl oligosaccharides coupled to proteins
T-dependent analogues of .alpha.(1,6)dextran. J. Immunol., 1989.
142: p. 863-870.
[0229] 14. Chen, H. T., S. D. Makover, and E. A. Kabat,
Immunochemical studies on monoclonal antibodies to
stearyl-isomaltotetraose from C58/J and a C57BL/10 nude mouse. Mol.
Immunol., 1987. 24: p. 333-338.
[0230] 15. Wang, D., et al., Two families of monoclonal antibodies
to .alpha.(1,6)dextran, V.sub.H19.1.2 and V.sub.H9.14.7, show
distinct patterns of J.sub.k and J.sub.H minigene usage and amino
acid substitutions in CDR3. J. Immunol., 1990. 145: p.
3002-3010.
[0231] Second Series of Experiments
[0232] I. Materials and Methods
[0233] A. Carbohydrate Antigens and Antibodies
[0234] Carbohydrate-containing macro-molecules applied in FIG. 7,
and anti-.alpha.(1,6) dextran mAbs, 16.4.12E (IgA/kappa)(6), 4.3F1
(IgG3/kappa)(5), and 45.21.1 (IgA/kappa)(8), were adopted from the
collection of the late Professor Elvin A. Kabat of Columbia
University. Purified proteins of 4.3F1, 45.21.1 and 16.4.12E were
obtained by a procedure of affinity purification as described(9).
The biotinylated or FITC-conjugated anti-dextran antibodies were
prepared in our laboratory following standard protocols(10). The
FITC-conjugated .alpha.(1,6) dextrans of moleuclar weight 20 kDa,
70 kDa, and 2,000 kDa, FITC-inulin, a biotinylated anti-human IgG
antibody, an alkaline phosphatase-conjugated anti-human IgM, and
streptavidin conjugates, were purchased from Sigma (St. Louis,
Mo.). Antibodies for cell type/lineage analysis, including
antibodies specific for murine CD11b/MAC1, MAC3, TCR-.alpha.,
TCR-.beta., CD3, CD4, CD5, CD8, CD19, B220, Syndecan-1, a mouse
IgG3 isotype standard (A12-3), and a streptavidin conjugate of
Texas Red, were from BD-PharMingen (San Diego, Calif.). A
streptavidin-Cy3 conjugate was purchased from Amersham Pharmacia
(Piscataway, N.J.), and a red fluorescence substrate of alkaline
phosphatase, Vector Red, from Molecular Probes, Inc. (Burlingame,
Calif.).
[0235] B. Printing Carbohydrate Microarrays
[0236] A high-precision robot designed to produce cDNA microarrays
(GMS 417 Arrayer; Genetic Microsystems, Inc., Woburn, Mass.) was
utilized to spot carbohydrate antigens onto the glass slides
precoated with nitrocellulose polymer (FAST Slides; Schleicher
& Schuell, Keene, N.H.). Carbohydrate antigens were dissolved
in saline (0.9% NaCl) in concentrations as specified in the Figure
legends. They were printed with spot sizes of .sup..about.150 .mu.m
and at 375-.mu.m intervals, center to center. The printed
carbohydrate microarrays were air-dried and stored at room
temperature without desiccant before application.
[0237] C. Staining and Scanning of Carbohydrate Microarrays
[0238] Immediately before use, the printed carbohydrate microarrays
were rinsed with PBS, pH 7.4, with 0.05% (vol/vol) Tween 20 and
then blocked by incubating the slides in 1% (wt/vol) BSA in PBS
containing 0.05% (wt/vol)NaN.sub.3 at 37.degree. C. for 30 minutes.
They were then incubated at room temperature with antibodies at an
indicated titration in 1% (wt/vol) BSA in PBS containing 0.05%
(wt/vol) NaN.sub.3 and 0.05% (vol/vol) Tween 20. Application of
secondary antibodies or streptavidin conjugates is specified in
figure legends. The stained slides were rinsed five times with PBS
with 0.05% (vol/vol) Tween 20, air-dried at room temperature, and
then scanned for fluorescent signals. The stained microarrays were
scanned with a ScanArray 5000 Standard Biochip Scanning System
(Packard BioChip Technologies, Inc., Billerica, Mass.) and data
analyzed using Quant Array version 2.1 software associated with the
system.
[0239] D. Elisa and in Situ Immunofluorescence
[0240] ELISA and immunofluorescence staining were carried out as
described (6, 11). The dextranase treatments were performed by a
preincubation of tissue sections with dextranase(Sigma) at 0.5
unit/ml in 100 mM potassium PBS, pH 6.0, 37.degree. C. for 60
minutes. This condition allows a complete removal of molecules of
FITC-.alpha.(1,6) dextran that were specifically trapped by immune
cells in the spleen sections of .alpha.(1,6) dextran-immunized mice
(11).
[0241] II. Results and Discussion
[0242] A model system for establishing carbohydrate microarray
technology. The dextrans and anti-dextran antibodies (1, 2) were
applied to establish methods for immobilizing carbohydrate polymers
on glass slides. Dextrans are polymers composed entirely of
glucose, produced mainly by bacteria of the family Lactobacillaceae
and of the genera Leuconostoc and Streptococcus. Dextran molecules
derived from different strains may, however, differ significantly
in their glycosidic linkage compositions. Unlike proteins that are
linked solely by a peptide bond, carbohydrates utilize many
possible glycosidic linkages so as to diversify their structures
extensively. Some dextran preparations are predominantly or solely
.alpha.(1,6)-linked, forming molecules with dominantly linear chain
structures; others are composed of multiple glycosidic linkages,
including .alpha.(1,6)-, .alpha.(1,3)-, .alpha.(1,2)-, and others,
generating heavily branched molecules(3) (Table 1). Previous
immunological studies (2, 4) have demonstrated that such structural
characteristics are detectable by antibodies specific for different
antigenic determinants or epitopes of dextran molecules. This
system is, therefore, suitable for developing methods for
immobilization of carbohydrate antigens and for investigating their
immunological properties in a surface-immobilized
configuration.
[0243] We applied the fluorescein isothiocyanate (FITC)-conjugated
polysaccharides as probes to investigate whether
nitrocellulose-coated glass slides can be used to immobilize
microspots of carbohydrate polymers without covalent conjugation.
FITC-.alpha.(1,6) dextran preparations of different molecular
weights and a structurally distinct polysaccharide, inulin, were
printed on the glass slides using a microprinting device to produce
a carbohydrate microarray (FIG. 6A). Their fluorescent signals were
then captured and quantified by a microscanning system that was
developed for scanning complementary DNA (cDNA) microarrays. By
analyzing the fluorescent intensities retained on the slides after
extensive washing, we demonstrated that dextran preparations
ranging from 20 kDa to 2,000 kDa and inulin of 3.3 kDa were all
stably immobilized on the nitrocellulose-coated slide without
chemical conjugation. The efficiency of their immobilization was,
however, significantly influenced by the molecular weight. The
larger dextran molecules were better retained than the smaller ones
(FIGS. 6A, B).
[0244] To investigate whether immobilized carbohydrate
macromolecules preserve their antigenic determinants or epitopes,
dextrtan preparations of different linkage compositions(3) and of
different rations of terminal to internal epitopes (1, 4) were
printed on nitrocellulose-coated glass slides. These preparations
include N279, displaying both internal linear and terminal
nonreducing end epitopes; B1299S, heavily branched and expressing
predominantly terminal epitopes; and LD7, a synthetic dextran
composed of 100% .alpha.(1,6)-linked internal linear chain
structure. The dextran microarrays were incubated with monoclonal
antibodies (mAbs) of defined specificities, either a groove-type
anti-.alpha.(1,6) dextran 4.3F1 (IgG3)(5) or a cavity-type
anti-.alpha.(1,6) dextran 16.412E (IgA)(6). The former recognizes
the internal linear chain of .alpha.(1,6) destrans; the latter is
specific for the terminal nonreducing end structure of the
polysaccharide. As shown in FIG. 7A (left), the groove-type mAb,
4.3F1 (refs 5,7) bound to the dextran preparations with
predominantly linear chain structures, N279 and LD7, but bound
poorly to the heavily branched .alpha.(1,6) dextran, B1299S. By
contrast, when the cavity-type mAb 16.4.12E (FIG. 7A, right) was
applied, it bound to the immobilized dextran preparations having
branches (N279 and B1299S) but not those with only internal linear
chain structure (LD7). These patterns of antigen-antibody
reactivities are characteristically identical to those recognized
by an ELISA binding assay (FIG. 7B) and other classical
quantitative immunoassays for either the groove-type (4, 5, 7) or
the cavity type (4, 6) of anti-dextran mAbs. We conclude,
therefore, that dextran molecules immobilized on a
nitrocellulose-coated glass slide have their immunological
properties well preserved. Both their nonreducing end structure,
recognized by the cavity-type anti-.alpha.(1,6) dextrans, and the
internal linear chain epitopes, bound by the groove-type
anti-.alpha.(1,6) dextrans, are displayed on the surface after
immobilization and are accessible to antibodies in an aqueous
solution.
[0245] References
[0246] 1. Wang, D. & Kabat, F. A. Carbohydrate antigens
(polysaccharides); Structure of Antigens, Vol. 3 (ed. M. H. V. V.
Regenmortal) 247-276 (CRC Press, Boca Raton Fla.; 1996).
[0247] 2. Wang, D. et al., The repertoire of antibodies to a single
antigenic determinant. Mol. Immunol. 28, 1387-1397 (1991).
[0248] 3. Jeanes, A. Immunochemical and related interactions with
dextrans reviewed in terms of improved structural information. Mol.
Immunol. 23, 999-1028 (1986).
[0249] 4. Cisar, J., Kabat, E. A., Dorner, M. M. & Liao, J.
Binding properties of immunoglobulin combining sites specific for
terminal or nonterminal antigenic determinants in dextran. J. Exp.
Med. 142, 435-459 (1975).
[0250] 5. Wang, D. et al., Two families of monoclonal antibodies to
.alpha.(1,6) dextran, V.sub.H19.1.2 and V.sub.H9.14.7, show
distinct patterns of J.sub.K and J.sub.H minigene usage and amino
acid substitutions in CDR3. J. Immunol. 145, 3002-3010 (1990).
[0251] 6. Matsuda, T. & Kabat, E. A. Variable region cDNA
sequences and antigen binding specificity of mouse monoclonal
antibodies to isomaltosyl oligosaccharides coupled to proteins.
T-dependent analogues of .alpha.(1,6)dextran. J. Immunol. 142,
863-870 (1989).
[0252] 7. Chen, H. T., Makover, S. D. & Kabat, E. A.
Immunochemical studies on monoclonal antibodies to
stearyl-isomaltotetraose from C58/J and a C57BL/10 nude mouse. Mol.
Immunol. 24, 333-338 (1987).
[0253] 8. Sharon, J., Kabat, E. A. & Morrison, S. L.
Association constants of hybridoma antibodies specific for
.alpha.(1.fwdarw.6) linked dextran determined by affinity
electrophoresis. Mol. Immunol. 19, 389-397 (1982).
[0254] 9. Lai, E. & Kabat, E. A. Immunochemical studies of
conjugates of isomaltosyl oligosaccharides to lipid: production and
characterization of mouse hybridoma antibodies specific for
stearyl-isomaltosyl oligosaccharides. Mol. Immunol. 22, 1021-1037
(1985).
[0255] 10. O'Shannessy, D. J., Dobersen, M. J. & Qarles, R. H.
A novel procedure for labeling immunoglobulins by conjugation to
oligosaccharide moieties. Immunol. Lett. 8, 273-277 (1984).
[0256] 11. Wang, D., Wells, S. M., Stall, A. M. & Kabat, E. A.
reaction of germinal centers in the T-cell-independent response tot
he bacterial polysaccharide .alpha.(1.fwdarw.6)dextran. Proc. Natl.
Acad. Sci. USA 91, 2502-2506 (1994).
[0257] Third Series of Experiments
[0258] I. Introduction
[0259] A microbial infection may expose and release multiple
antigenic substances to a host, eliciting a comprehensive host
immune response, including a B cell response, which produces
specific antibodies, and a T cell response, resulting in specific T
cell activation and cytokine production. There is also an
activation of macrophage, dendritic cells and other accessory
cells, leading to production of differential profiles of cytokines
and inflammation factors. Some microbial substances are lethal to a
host upon their immediate release or after interacting with host
cells or cellular factors. The interplay of different types of host
cells and multiple protein factors and their interaction with the
invading pathogen determine the progress and consequence of a
microbial infection. For example, in an anthrax infection, the
pathogen Bacillus anthracis may expose and release a number of
antigens of distinct structural characteristics to a host,
eliciting a comprehensive picture of a host response (FIG. 8).
[0260] To better understand the pathogenesis mechanisms of an
infectious disease, it is crucially important to monitor the
full-spectrum of the multi-parameter host responses and identify
the characteristic patterns of the response. Monitoring such a
complex host response and host-microbe interaction has long been a
challenge to biomedical scientists and clinicians. Many
antigen-antibody binding assays are currently in use for clinical
diagnosis of infectious and non-infectious diseases. These include
the classical direct immunoassays, such as, immunodiffusion,
immunoelectrophoresis, agglutination and immunoprecipitation, and
recently developed methods, including immunofluorescence,
radioimmunoassay (RIA), enzyme-immunoassay (EIA) and western blot.
These approaches take advantage of the specificity of
antigen-antibody interaction but are designed to operate on a
one-by-one basis.
[0261] Rapid progress of the genome sequencing projects has led to
the development of a generation of high throughput technologies for
biological and medical research. These include the nucleic
acid-based microarrays (6,7) or DNA chips (8), and the
protein-based microarrays (9,10). Our recent efforts have been
focusing on the development of a carbohydrate and protein-based
microarray technology to extend the scope of biomedical research on
carbohydrate-mediated molecular recognition and anti-infection
responses (see reference 11 and next section for our most recent
progress).
[0262] The microspot format of surface displaying biological
molecules has the advantage of achieving a highly sensitive and
simultaneous detection of multiple binding partners in solution.
Since the amount of molecule in the solution phase that is required
for saturating the surface immobilized microspots of molecules is
considerably small, binding can be achieved with a relatively lower
molar concentration of molecules in solution. In brief, it is
believed that the smaller microspot is better than the bigger spot
in its sensitivity of detection in an assay system (12,13).
[0263] (A) Category a Pathogens and Their Genomic Information
[0264] High-priority infectious agents that pose current risks to
our national security include multiple microbial pathogens known as
Category A pathogens, such as Bacillus anthracis (anthrax),
Clostridium botulinum (botulism), Yersinia pestis (plague), Variola
viruse (smallpox), Francisella tularensis (tularemia) and viral
hemorrhagic fevers (Lassa virus, Ebola virus, Marburg virus and
Lymphocytic Choriomeningitis virus). An assay system that allows
detection and characterization of these pathogens in a single assay
using limited clinical specimens is currently unavailable.
[0265] A great deal of genomic information is available for
numerous pathogens (14-16), providing many opportunities to the
field of infectious disease research. The identification of
candidates from a large repertoire of genes, including a large
panel of genes whose function is unknown, for developing diagnostic
protein biochips is, however, a current challenge. As detailed
below, we have established a strategy to facilitate the selection
and identification of genes that have potential to serve as novel
molecular targets for vaccination and diagnosis. Information
provided by whole genome sequencing of microbes may also boost
progress in the molecular engineering of artificial species of
desired characteristics. There is also concern that the release of
laboratory amplified genetic material to the environment may
facilitate the natural occurrence of recombinant or newly evolved
microbial species or strains. Such emerging microbes could be
either beneficial or harmful to public health depending on the
nature of the microorganism as well as the way by which they are
utilized. Therefore, in designing a high throughput immunoassay one
must consider the detection of existing pathogens as well as their
mutants or recombinant forms. This goal is achievable if a panel of
antigens of a given microbe and/or their specific antibodies are
applied to produce a diagnostic biochip to detect multiple
molecular targets of the pathogen. Applicants have tested the
feasibility of this proposal by printing a panel of HIV-1 proteins
on a single chip to monitor the HIV-1 infections by different
classes or clades of HIV-1 viruses (see FIGS. 11 and 12).
[0266] (B) Antigen-Based Biochip
[0267] An antigen biochip is designed to detect and quantify
antibodies in body fluids. During infection, whether viral,
bacterial, fungal, or parasitic, the host usually responds with
formation of antibodies, which can be detected by modification of
any of the methods used for antigen detection. The formation of
antibodies and their time course depends on the antigenic
stimulation provided by the infection. Recognition of these
patterns provides us with evidence of recent or past infection.
Establishing a biochip to display a large panel of microbial
antigens and the host derived autoantigens will substantially
extend the scope of biomedical research on the biological
relationship between host and microorganism, as well as
understanding the molecular mechanisms of infectious diseases. Both
carbohydrate antigens and protein antigens can be of importance for
the design and production of a diagnostic biochip.
[0268] Carbohydrate structures of microbial origin, including
polysaccharides, glycolipids and glycoproteins, frequently serve as
the main antigenic structures to which host cells recognize and
mount a response (17). A single microbial antigen may, however,
display multiple antigenic determinants, with one or a few of them
predominating in a given infection. For example, a relatively
simple microbial polysaccharide, .alpha.(1,6) dextran N279,
displays both the internal linear chain and terminal non-reducing
end structures as distinct antigenic determinants. When the
polysaccharide was injected to a Balb/c mouse, it elicited a
predominant antibody response directed to the internal linear chain
epitope of .alpha.(1,6)dextran. The internal linear chain epitope
is thus defined as a dominant antigenic determinant to the host.
The antibodies elicited were solely or dominantly of the
groove-type anti-dextran. In this case, the terminal non-reducing
end structure is apparently the minor antigenic determinant. A
dominant antigenic determinant should, therefore, be placed as a
high priority to serve as a target for vaccine design or for
developing an immunoassay for diagnosis.
[0269] Identifying a minor antigenic determinant is, however, also
important since there is the possibility that a dominant antigenic
structure is not suitable for vaccination or diagnostic
application; whereas, a minor antigenic determinant may serve as a
molecular target for these applications. Some microbial antigens
may share or mimic host components. This is known as antigenic
cross-reactivity. For examples, .alpha.(1.fwdarw.8)NeuNAc in the
capsule of the group B meningococcus and of E.coli K1 is found on
glycoproteins and gangliosides--of human tissues, complicating the
application of the capsular polysaccharides for vaccination,
especially in infants who have expression of polymeric forms of the
carbohydrate structure. There are also circumstances that a
dominant antigenic structure can be useful for diagnosis but not
for vaccination. For example, the Gag p24 protein of HIV-1 elicits
a dominant antibody response in most AIDS patients. Detecting
anti-Gag antibodies has diagnostic value. These antibodies,
however, have no HIV-1 neutralization activity since Gag is not
expressed on the surface of the HIV-1 virus. Even for the envelope
protein, gp120 of HIV-1, which is surface displayed and accessible
to antibody recognition, there are only a few epitopes of the
glycoprotein to which antibodies can be effective in virus
neutralization (18).
[0270] With current immunological advances, converting a minor
antigenic determinant into a dominant one is technically
achievable. For example, the antigen may be coupled with a carrier
molecule, forming a highly antigenic conjugate molecule. As
demonstrated with a model antigenic system .alpha.(1,6)dextran, an
isomaltotriose (IM3), which is derived from the polysaccharide
.alpha.(1,6)dextran, can be coupled to BSA or KLH to produce a
semi-synthetic glycoprotein to display the minor antigenic
determinant of the polysaccharide, i.e., the terminal non-reducing
end epitopes of .alpha.(1,6)dextran. When this glycoconjugate was
injected to Balb/c, a dominant antibody response to the terminal
antigenic determinant is induced. Immunization with such an
artificial semi-synthetic glycoconjugate is now well known as the
conjugate vaccine. A notable example is vaccination with the
protein-conjugates of Haemophilus influenza type b polysaccharide
that resulted in the decline in the incidences of H. influenza
meningitis and other infections in infants and children (19,20).
The presence of polysaccharides and glycoproteins in Bacillus
anthracis has been recognized for some time (4, 21,22). Whether
these carbohydrate structures are suitable targets for anthrax
vaccination is, however, an open question.
[0271] Recognition of cross-reactive antigenic determinants is
frequently shown to be of biological and medical significance.
There are a number of documented cases in which microbial antigens
mimic the structures of host components, assisting a microbe to
escape from a host's immune defense (23,26). Such mimicking
microbial antigens may also induce autoimmune disorders and
contribute to the pathogenesis of an infectious disease (23,27).
Identification and characterization of such antigenic structures
may lead to a better understanding of the molecular mechanisms of
infectious diseases.
[0272] Antigenic structures that are not expressed on the surface
of a microorganism also have an important diagnostic value. For
example, detection of antibodies that are specific for the surface
antigen of hepatitis B (HBsAg) may indicate an early viral
infection or a successful vaccination; detecting antibody
specificities to multiple viral antigens, such as surface antigen
(HBsAg) plus core antigen (HBcAg) or HBsAg plus a relevant e
antigen (HBeAg) suggests an active infection or the progression of
the disease (28,29). A microbial pathogen, either a virus or a
bacterium, may release multiple antigenic substances that trigger
host antibody responses. In an early infection, as an initial
immune response, the antibodies elicited are mainly those bound to
the surface antigens and are predominantly IgM antibodies. With the
progression of an infectious disease, for example in an AIDS
patient a few months post-serum conversion, a large panel of IgG
antibodies with different specificities, including anti-gp120
(envelope protein) and anti-Gag p55 polyproteins of HIV-1, can be
detected in the serum of the patient. Thus, applying a combination
of surface and non-surface antigens on a biochip for diagnosis can
assist in the recognition of the stages or steps of an infection
and provide information to predict disease progression and to
evaluate the efficacy of a therapeutic agent or strategy.
[0273] Development of drugs or therapeutic strategies against
microbial infection requires a better understanding of the
pathogenic mechanisms of an infectious disease. For example, the
anthrax toxin is lethal upon its activation. This process involves
multiple steps of molecular and cellular interactions of anthrax
proteins, host cells and protein factors (1-3). The protective
antigen of anthrax, named PA, is an integrated component of the
lethal toxin of Bacillus anthracis. It binds to a specific cellular
receptor and forms toxic, cell bound complexes with edema factor
(EF) and lethal factor (LF) (1-3). This understanding leads to the
development of a polyvalent inhibitor-based therapeutic strategy to
protect a host from lethal attack by the toxin (5). Technically, it
is of crucial importance to identify the structural moieties or
epitopes that play key roles in forming the toxic complex or in the
interaction of PA and its macrophage receptor. Generally speaking,
the epitopes of a protein or a glycomer that interact with or are
recognized by their partners are surface exposed and are,
therefore, recognizable by specific antibodies. Identifying such
antigen-antibody pairs is of crucial importance. Such pairs could
serve as specific probes for the screening of smaller molecules to
identify drug candidates that may block the effect of the anthrax
toxin. Development of an antigenic structure-based biochip of large
capacity and diversity would facilitate efforts to identify such
key structural elements and screen for their specific antibodies
for drug development.
[0274] Thus, for the purpose of diagnosis, vaccination and drug
development, it is important to recognize the dominant and minor
antigenic determinants for a specific anti-infection antibody
response to characterize the specificity and cross-reactivity of an
antigenic structure; and to have a full-panel scanning of the
repertoires of antibody specificities elicited by an infection.
Such investigation has been impossible owing to a lack of a high
throughput, multi-parameter assay system. Developing a carbohydrate
and protein-based biochip to present a large repertoire of
antigenic structures, including those displaying a single antigenic
determinant on each microspot, would substantially facilitate these
investigations.
[0275] (C) Antibody-Based Biochip
[0276] In principle, an antigen-based biochip is designed to detect
and quantify specific antibodies and therefore, diagnose an
infection. This method is not for the detection and quantification
of antigens that are released from an infectious agent. Detecting
microbial antigens in serum or other body fluids is generally more
difficult than detecting antibodies. The former has, however,
higher diagnostic value than the latter. It is, therefore,
necessary to establish a highly sensitive antibody-based microarray
to detect microbial antigens.
[0277] The complexity of a host anti-infection response further
challenges the development of a multi-parameter immunoassay for
characterizing infectious diseases. During an infection, a large
panel of antigenic substances of distinct structural
characteristic, including protein, polysaccharide, glycolipid,
glycoproteins and nucleic acid, may be released to trigger the host
immune response. These substances may differ significantly in their
immunological properties and, therefore, elicit characteristic
patterns of host responses. For example, protein antigens fail to
elicit antibody responses in mice lacking a thymus but
polysaccharides and other macromolecules with repetitive antigenic
determinants can induce unimpaired antibody responses in these mice
(30-32). The former are termed T-dependent (TD) antigens and the
latter are T-independent (TI) antigens. The differences of TI- and
TD-antigenic responses, which are recognizable by in vitro
immunoassay, include a series of humoral factors, such as antigen
specific antibodies, their Ig-isotypes, cytokines and other
inflammation factors.
[0278] To better understand infectious disease, the full spectrum
of anti-infection responses must be studied. These include specific
antibody responses as well as the antigen-non-specific cytokine
responses and other host responses. Cytokines are soluble proteins
or glycoproteins, which play a critical role in controlling
development or differentiation of lymphocytes and in regulating
their anti-infection responses. For example, a microbial infection
or vaccination may activate certain sub-types of T cells, either T
helper 1 (Th1) cells or T helper 2 (Th2) cells. These specialized
Th cells can produce unique profiles of cytotokines. Th1 secrete
IL-2, IL-3, TNF-.alpha. and IFN-.gamma.; Th2 secrete IL-3, IL-4,
IL-5, IL-6, IL-9, IL-10 and some TNF-.alpha.. For anti-infection
responses, induction of IgA-antibodies at mucosal sites is of
critical importance. T cells and their cytokines have been shown to
play a critical role in various stages of IgA response (33-35),
including induction of Ig class switching of IgM to IgA and of
terminal differentiation of IgA-committed B cells. Many researchers
believe that IgA responses are highly Th-2 dependent (36-39), since
there is evidence that Th2 cytokines, such as IL-4, IL-5 and IL-6,
are required to induce terminal differentiation of IgA-committed B
cells. By contrast, the Th1 cytokine, IFN-.gamma., antagonizes IL-4
in its IgA induction.
[0279] Our current efforts focus on the development of high
throughput post-genomic technologies to extend the scope of
biomedical research on human infectious diseases and the human
immune response. A carbohydrate and protein-based microarray has
been prepared, making it possible to display a large collection of
antigens on a single biochip for probing the repertoires of
antibody specificities and for studying carbohydrate and protein
mediated molecular recognition on a large scale. This microarray
platform has achieved the sensitivity to detect a broad range of
human antibodies with as little as a few microliters of serum
specimens and has reached the capacity to include antigenic
preparations of most common pathogens. This technology is,
therefore, readily applicable for large-scale production of an
antigen/antibody microarray. In this invention, we describe the
establishment of a detailed procedure for the industrial scale
production of diagnostic biochips to enable simultaneous detection
and characterization of a wide range of microbial infections, which
include all listed Category A biological warfare agents. These
include (a) the design and production of a carbohydrate-based
microarray composed of microbial polysaccharides, its derivatives
and a large panel of carbohydrate-containing macromolecules of
distinct sugar structure; (b) a procedure to take advantage of the
genomic information available for specific pathogens to design and
produce proteomic microarrays to identify novel molecular targets
for vaccination, diagnosis and drug development; (c) a method to
produce an antibody-based microarray system to monitor microbial
antigens in vivo and in vitro and to enable full-panel scanning of
cytokines and other inflammation factors with limited amount of
clinical specimens; and (d) the design and production of highly
sensitive diagnostic biochips, including Diagnostic Biochip A and B
for the rapid detection of all the biological warfare agents listed
by the CDC as Category A pathogens and Diagnostic Biochip C for
simultaneous detection and characterization of a wide range of
common infectious diseases, which are caused by about 200 human
pathogens.
[0280] II. Methods to Produce Large Capacity and High-Density
Biochips (.sup..about.30.000 Microspots/Microchip)
[0281] Compared with DNA- and protein-based microarrays, the
carbohydrate microarray technology has certain technical advantages
and disadvantages. An obvious advantage is that the purified
polysaccharides are generally stable in various conditions, either
as dried solid or in aqueous solution, at room temperature or at
4.degree. C. in storage. Unlike protein microarrays, wherein
protein-denaturing and/or conformational alteration in microarray
printing and storage is a major challenge to the technology, there
is no such serious concern for most (if not all) carbohydrate
antigens. A disadvantage of the carbohydrate microarray technology
is that methods for high throughput production of carbohydrate
macromolecules or complex carbohydrates, which is equivalent to the
PCR for amplifying DNA or the cloning and expression method for
producing proteins, are yet to be developed. Classical methods to
obtain pure carbohydrate antigens include (a) isolation and
purification from biological materials, such as cells, tissues or
biological fluids; (b) chemical synthesis; and (c) enzymatic in
vitro synthesis. A Rapid progress in establishing high throughput
technology for the in vitro synthesis of carbohydrate
macromolecules or complex carbohydrate molecules is most likely
unexpected. The availability of purified carbohydrate antigens is,
therefore, a potential rate-limiting factor to the carbohydrate
microarray industry.
[0282] Our optimized methods for printing microarrays allows one:
(i) to reduce amount of antigen needed for microarray printing;
(ii) to increase the detection sensitivity and biochip capacity;
(iii) to reduce the time that is necessary for a microarray
printing cycle; (iv) to prevent cross-contaminations in printing
microarrays; and (v) to reduce the cost of producing a diagnostic
biochip substantially.
[0283] (A) Printing Microspots of Smaller Sizes and
High-Densities
[0284] The general printing procedure is schematically illustrated
in FIG. 9. A high-precision robot designed to produce cDNA
microarrays was utilized to spot carbohydrate antigens onto glass
slides that were pre-coated with nitrocellulose polymer. In
addition, the following were also used:
[0285] a) STEALTH 3 pins for printing microspots 150 microns in
diameter on nitrocellulose slides. This allows the arraying of
10,368 spots per FAST-slide with spot intervals of 250micron,
center-to-center; and
[0286] b) STEALTH 2.5 pins for printing spots at about 100 microns
in diameter on the surface. 28,800 spots can be patterned on a
single FAST slide. If each antigen preparation is printed as an
array of four identical microspots, (See FIG. 12 below), 7,200
antigenic molecules can be included on a single slide. This system
has, therefore, enough capacity to include most known human
microbialpathogens and tumor-associated antigens.
[0287] III. Diagnostic Biochips
[0288] (A) Diagnostic Biochip A
[0289] This biochip is designed to enable the simultaneous
detection of all Category A infectious diseases. In its clinical
application, two-step staining is required. A number of carefully
selected pairs of antigen and antibody are be printed on slides for
each pathogen. The immobilized antigen serves as a probe to detect
antibody in solution; the surface displayed antibody is used to
capture a specific antigen in the solvent. This assay is,
therefore, capable of detection of both antigens and antibodies in
a single assay. Since an immobilized antigen may display multiple
antigenic determinants, a microspot of antigen may capture
antibodies that recognize different antigenic determinants
expressed by the antigenic molecule. This makes the biochip system
highly sensitive. Its detection specificity is, however, at the
level of the antigenic molecule but not at the level of a single
antigenic determinant. In case an antigen expresses a
cross-reactive antigenic determinant, its detection-specificity
will be reduced (See the design of diagnostic biochip B for further
discussion).
[0290] (1) 8-Chamber Sub-Arrays (See FIG. 10):
[0291] Each micro-glass slide contains eight well-separated
sub-arrays of identical contents. There are 600 microspots per
sub-array, with spot sizes of approximately 200 micron at 300
micron intervals, center-to-center. A single slide is, therefore,
designed to enable eight detections;
[0292] (2) Contains:
[0293] (a) Microspots: Each 600-spot sub-array is composed of
carbohydrate and protein antigens, as well as antibodies specific
to microbial antigens. The immobilized antigens will allow for the
detection of human antibodies elicited by an infection; whereas
immobilized antibodies are for the detection of microbial
antigens;
[0294] (b) Repeats and dilutions: Each antigen/antibody will be
printed at 0.5-1.0 mg /ml and at one to ten dilution of the initial
concentration for the second concentration. Each preparation at a
given concentration will be repeated three times; and
[0295] (c) Antibody isotype standard curves: Human antibodies of
IgG, IgA and IgM isotype of known concentrations will serve as
standard curves for antibody detection and normalization.
[0296] (3) Assay Mechanisms:
[0297] (a) Detection of antibodies: use of immobilized antigens to
capture antibodies in solution, which are then recognized by the
tagged anti-human antibodies. In principle, this is an indirect
immunoassay;
[0298] (b) Detection of microbial antigens: use of immobilized
antibodies to capture antigens in solution and application of the
tagged antibodies specific for the corresponding antigens to
identify the captured antigens. This is known as a "Sandwich"
immunoassay.
[0299] (4) Application and Staining Procedure:
[0300] Use of 0.5-1.0 microliters of serum specimens for the
detection of a) specific antibodies in body fluids and b) presence
of antigens in vivo and in vitro. Two steps of staining and
approximately 5 hrs is required to complete the biochip
analysis.
[0301] (5) Specificity & Sensitivity:
[0302] a highly sensitive biochip system with specificity at the
level of an antigenic molecule.
[0303] (B) Diagnostic Biochip B
[0304] This biochip is designed to enable the rapid diagnosis of
the Category A infectious diseases, which requires only a single
step of staining in clinical diagnosis. The time required for a
biochip assay is, therefore, substantially shorter. Competitive
immunoassay is used for rapid diagnosis. A competitive immunoassay
requires an antigen/antibody pair. Either antigen or antibody can
be immobilized on the solid surface, which will then interact with
an antigen or antibody in solution. The immobilized antigen and
tagged antibody in solution forms a specific probe to detect both
antigen and antibody in clinical specimens. The free-antigen or
antibody competitively inhibits the binding of the tagged antibody
to the antigen immobilized on the solid surface. Key
characteristics of this "competitive" one-step biochip that differ
from Diagnostic Biochip A are summarized below:
[0305] (1) Chip Contents:
[0306] A panel of carefully selected microbial antigens, whose
specific antibodies are available, will be printed on the chip with
the method described above.
[0307] (2) Assay Mechanisms:
[0308] The free antigens or antibodies can compete with the labeled
antigen or antibody to bind the immobilized antigens or antibodies.
Given that antibodies are generally more stable than other protein
molecules in solution and are suitable for standardized production,
printing antigen on a series of glass slides and using
fluorescent-tagged antibodies for staining is preferred. If a
clinical specimen in question contains a specific antibody for the
target antigenic determinant or the antigen, the binding of the
tagged antibody to the antigen will be competitively inhibited.
[0309] (3) Application and Staining Procedure:
[0310] One-step staining of the biochip allows detection of
specific antibodies in body fluids of antigens in vivo and in
vitro. The time required for biochip analysis, is therefore,
reduced by 4 hours.
[0311] (4) Specificity & Sensitivity:
[0312] Diagnostic Biochip B is a highly sensitive and specific
biochip system. Its specificity is at the level of a single
antigenic determinant.
[0313] (C) Diagnostic Biochip C
[0314] Diagnostic Biochip C is composed of a large repertoire of
carbohydrate and protein antigens as well as antibodies. This is a
largely extended Diagnostic Biochip A. This biochip makes it
possible to diagnose and characterize a wide range of microbial
infections using a few microliters of serum specimen. This
microarray has the printing capacity to include most common
pathogens. Practically, it is limited by the availability of
specific antigen preparations and their antibodies. 1000-2000
distinct antigens and antibodies can be printed on the chip. This
enables a simultaneous detection of about 300 microbes, which
include about 50-100 human pathogens. General properties of the
master microarray are summarized as follows:
[0315] (1) Capacity:
[0316] 15,000-20,000 microspots per micro-glass slide, with spot
sizes approximately 150 micron in diameter and at 200-micron
intervals, center-to-center.
[0317] (2) Diversity:
[0318] 1000-1500 distinct antigenic preparations, about 100
antibodies specific for microbial antigens and about 30-50
antibodies for detecting human cytokines and other inflammation
factors.
[0319] (3) Repeats and Dilutions:
[0320] Each antigen has four dilutions (0.5 mg/ml to begin with)
and each dilution has three repeats.
[0321] (4) Antibody Standard Curves:
[0322] Human antibodies of IgG, IgA and IgM isotype of known
concentrations are printed on the chip to produce standard curves
for normalization across the experiments and for quantitative
calculation of the titer of the specific antibodies of a given
isotype captured by the carbohydrate antigen.
[0323] (D) Diagnostic Biochip D
[0324] This biochip is composed of about 4,000 microspots of
antigen and antibody preparations. Each antigen or antibody
preparation of a given dilution was printed with four repeats in a
vertical line of microspots on the biochip. This allows us to
visually observe and statistically analyze the reproducibility of
microarray printing and staining. The significance and sensitivity
of antibody detection for each antigen at a given antigen
concentration can also be statistically calculated. Some
preparations, for example the gp120 glycoprotein of HIV-1 as
highlighted in a square in FIG. 11, were printed from left to right
in a series dilution of one to five, beginning at 0.5-1.0 mg/ml and
with four dilutions thereafter. These biochips were stained with
the normal or HIV-1 infected human serum specimens with the methods
described above. Pictures of the ScanArray visualization of the
multi-color fluorescent staining of biochips are shown in FIG.
11.
[0325] In this experiment, we have demonstrated that: (a) a
procedure for microarray printing, antigen/antibody immobilization
and antibody staining is precisely reproducible. On the same
biochip, the correlation factors of data crossing different
microspots of the same preparation is in the range of (0.98) to
(1.00); (b) detection of human serum antibodies by this method is
highly sensitive. A few microliters of serum specimen allows a
full-panel scanning of the repertoires of human antibodies of
different Ig-isoty{acute over (p)}es in a single assay. A large
repertoire of antibody specificities were recognized in the normal
and HIV infected individuals; (c) the specificity of this system is
illustrated by recognizing the epitope-binding specificities of
monoclonal anti-dextran antibodies and by specific detection of
human serum antibodies for the gp120 glycoproteins and gag p24 of
HIV-1 in AIDS patients; (d) a large repertoire of microbial
antigens can be patterned on a single micro-glass slide, reaching
the capacity to include most common and conditional pathogens; and
(e) a biochip-based high throughput technology requires a strong
bioinformatic presence to support it. While one may need a day to
perform a biochip assay for a given clinical specimen, one may need
many days to process the large amount of data produced by the
biochip analysis. Developing advanced computer algorithms to
facilitate this process is of equal importance with improving the
hardware of the biochip technology.
[0326] (E) Diagnostic Biochip E
[0327] This is a proteomic microarray-based biochip produced with
the microarray platform used to produce carbohydrate microarrays.
Discovery of a large repertoire of genes in the genome-of
microorganisms has provided novel targets for vaccination,
diagnosis, and drug development against microbial infection. To
take advantage of this, HIV-1, whose genome is relatively small and
has been completely sequenced, is used herein to establish the
protein-based microarray technology. A large panel of purified HIV
proteins, including gp120 proteins derived from different clades of
HIV-1, Gag p55, p24, P6, P7, P17 and/or their E. coli expressed
GST-fusion proteins, Tat, Nef, Integrase, and reverse transcriptase
(RT), were printed and immobilized on the chemically modified glass
slides. These protein chips were then applied to probe antibodies
in normal and AIDS patients. As shown in FIG. 12, most HIV proteins
printed on the chip gave positive detection of antibodies in
HIV-infected individuals but not in normal controls, showing the
sensitivity and specificity of this protein chip.
[0328] Therefore, we conclude that: (a) these positively stained
HIV proteins were stably immobilized on the biochip and retained
their immunological properties; (b) the protein microarray system
has reached the sensitivity to probe the repertoires of antibodies
in clinical specimens; and, most importantly, (c) it is feasible to
introduce a large panel of protein products of genes, including
both surface and non-surface proteins, and newly discovered genes,
for our biochip production and diagnostic application. This finding
is significantly important since it allows us to explore the
application of genomic information and genetic material that has
been accumulated by the genome projects.
[0329] (F) Diagnostic Biochip F
[0330] This is an antibody-based microarray and is designed to
enable a full-panel scanning of human cytokine. Cytokine detection
at the protein level is technically difficult. These proteins are
potent biological agents that function at very low concentrations
(7). Some cytokine molecules are unstable after secretion or
activation (7). A highly sensitive specific assay is required to
detect these molecules. A method currently used for cytokine
detection, either ELISA-based assays or radioimmunoassays, takes
advantage of the Sandwich-antibody assay. Specifically, the first
cytokine-specific antibody is immobilized on a solid surface to
capture cytokine in solution and then the second anti-cytokine
antibody is applied to detect the surface-immobilized cytokine. The
second anti-cytokine can be biotinylated, allowing signal
amplification with a labeled Streptavidin. These methods are highly
sensitive but limited in detecting a cytokine on a one-by-one
base.
[0331] We have extended our microarray platform to produce the
antibody-based biochips. The arrayed antigens allow specific
capture of antibodies from body fluids and the antibodies
immobilized on the glass chip can be used to detect soluble
antigens and host factors, such as cytokines and other inflammatory
factors. We immobilized a panel of anti-dextran mAbs at a
concentration of 0.5 mg/ml on a set of glass slides. These include
nitrocellulose-coated glass slides, polylysine treated slides,
silane-treated glass slides and untreated, precleaned glass slides.
These slides were then reacted with the fluorescence-tagged dextran
preparations of distinct structures. Only the nitrocellulose-slides
showed spots of specific fluorescent signals. Thus, anti-dextran
mAbs were immobilized on the nitrocellulose-glass slide and
retained their antigen binding specificities.
[0332] We further addressed whether the nitrocellulose-coated slide
can be applied to immobilize antibodies on a chip for long-term
storage. To investigate the condition for the preservation of
antibody microarrays, we placed these microarrays under different
conditions: (a) in an air-dried condition at room temperature; and
(b) in a blocking solution, stored at 4.degree. C. After six
months, we stained these antibody microarrays with tagged antigens.
We found that both group (a) and group (b) slides preserved
antigen-capturing activities and specificities. Signals obtained by
the two groups differ, however, significantly, with the latter much
stronger than the former. These experiments have indicated the
potential of the nitrocellulose-based non-chemical immobilization
method for antibody microarray production.
[0333] An antibody microarray for a full-panel scanning of human
cytokines was prepared. Polyclonal and monoclonal antibodies
specific for human cytokines are currently available through
various resources, providing a strong base for developing an
antibody microarray to enable a highly sensitive, high throughput,
full-panel cytokine scan.
[0334] We have found that the current microarray platform favors
detection of IgG antibodies but not IgM antibodies in solution. The
former is an immunoglobulin (Ig) monomer and the latter is an Ig
pentamer. One theory for this result is that the pore size of FAST
slides is too small to allow IgM antibody to enter and bind antigen
immobilized in the deeper layer of the 3D nitrocellulose network
structure (see ref. 40 for a 3D illustration of the nitrocellulose
coating). This finding is of significance since it provides
information for the development of an improved surface for
producing a diagnostic biochip.
[0335] IV. Methods to Search and Identify Novel Carbohydrate
Targets for Diagnosis and Vaccination
[0336] A large collection of carbohydrate containing macromolecules
are printed on the nitrocellulose-coated slides and then the
carbohydrate microarrays are stained with antibody preparations or
lectins to probe the microspots that display the antigenic
determinants in question. The monoclonal and polyclonal antibodies
elicited by a microbial antigen or by a pathogen, as well as serum
specimens of an infected individual are all useful reagents for
these analyses. A positively stained microspot indicates the
presence of a target molecule or antigenic determinant. Further
characterization of this antigen preparation could lead to the
identification of a suitable diagnostic molecule for a given
infectious disease.
[0337] We have employed B. anthracis as a model pathogen to
illustrate our research approach, from the identification of a
target molecule to the further characterization of the structure
using our carbohydrate microarray technology.
[0338] (A) B. anthracis Related Carbohydrate Antigens as Diagnostic
Molecular Targets
[0339] Carbohydrate structures present in B. anthracis are
potential molecular targets for vaccine development and for
diagnostic application. These carbohydrate structures include
glycoproteins and polysaccharides. Glycoproteins are expressed by
the dormant spore of B. anthracis and are recognizable by specific
lectins, such as Glycine max (41). Since the spore surface
interacts with the host initially in an anthrax infection, these
structures are likely the targets for host recognition and antibody
responses. They are therefore important for both vaccination and
diagnosis. Polysaccharides are expressed by the germinating spores
of B. anthracis, which are detected by monoclonal antibodies raised
against the cell wall Gal-NAG polysaccharide of the pathogen (22).
The exo- and/or cell wall Gal-NAG polysaccharide, which is present
in the culture medium for growing B. anthracis (4), can be isolated
from the cell wall of the microbe (22,42). Given that the Gal-NAG
polysaccharide is universally present among and specific for
strains of Bacillus anthracis (22), its presence in solution and
its structural stability, makes this molecule a potential target
for the identification of B. anthracis and for developing an
anthrax vaccine.
[0340] Early studies were conducted by the late Professors Michael
Heidelberger (see ref. 43 page 451 for a description) and Elvin A.
Kabat (44-46) to investigate the antigenic cross-reactivities among
blood group substances (ABO), Pneumococcus type 14-polysaccharide
and the cell wall polysaccharide of B. anthracis. A preparation of
the anthrax polysaccharide is available in the KABAT collection at
Columbia University. The anthrax polysaccharide was tested on a
carbohydrate microarray and confirmed that the preparation has
preserved its cross-reactivity to an anti-Pneumococcus type
14-polysaccharide antibody (data not shown).
[0341] Following are some important characteristics of the cell
wall Gal-NAG polysaccharide: (a) it is 12,000 Da in molecular
weight and contains galactose, N-acetylglucosamine, and
N-acetylmannosamine in an approximate molar ratio of 3:2:1 (22) or
10:2:1. This slight difference in molar ratio is attributed to
differences in the hydrolysis conditions applied (47); (b) it is
pyruvylated although the sugar residue and position of pyruvylation
are yet to be determined(47); (c) it is most likely B. anthracis
specific and expressed by both germinating spores and the
vegetative bacillus (22); (d) it shows antigenic determinants which
are cross-reactive to Pneumococcus type 14 polysaccharide (43); (e)
it has no human blood group substances A, B, or H activity (43);
and (f) anti-serum elicited by the anthrax polysaccharide has,
however, cross-reactivities with a preparation of hydrolyzed blood
group substance A that contains the type II Gal-GlcNAc sequences
(Gal .beta.1.fwdarw.4GlcNAc) (see ref. 43, page 452 for a
description of the study by Ivanovicx (1940)).
[0342] These characteristics lead to the conclusion that: (i) The
Gal-NAG polysaccharide expresses potent antigenic determinants to
mice and humans since the pyruvylated Gal-NAG structure is not
present in the host; (ii) the polysaccharide itself is, however,
poorly antigenic in solution since it is a T-independent antigen
with relatively low molecular weight (1.2 KDa). Given this
consideration, early observations that the polysaccharide was not
protective to animals challenged by B. anthracis can be attributed
to the nature of poor immunogenicity in its native configuration;
(iii) its Gal-NAG core structure is similar to a core structure of
blood group type II chain, Gal .beta.1.fwdarw.4GlcNAc. Pyruvylation
of the backbone structure blocks, however, its blood group
substance H reactivity; and (iv) given that the pyruvate group is
not seen in Pneumococcus type 14 polysaccharide, the anthrax
polysaccharide may contain a non-pyruvylated antigenic determinant
mimicking those of the Pneumococcus type 14 polysaccharide. This
leads us to conclude, therefore, that the cell wall polysaccharide
of B. anthracis contains more than one antigenic determinant.
[0343] A carbohydrate microarray was designed to maximize the
surface display of its potential antigenic structures using the
following method: (a) print the polysaccharide on a microchip to
display its antigenic determinants in the native configuration; (b)
isolate the pyruvylated and non-pyruvylated fractions of the
polysaccharide and apply them on the microarray as described by
Mesnage et al. (47). This allows investigation of the potential
dominant role of the pyruvylated sugar structure in its
antigenicity; and (c) synthesize a panel of oligosaccharide-protein
conjugates. In the chip design, we have also included structures of
oligosaccharides that are identical to or derived from those of
Pneumococcus type 14 polysaccharide and those reactive to lectin
Glycine max (specific for alpha-D-galactose or 2-acetamido
2-deoxy-alpha-D-galactose residues) that recognize a proposed spore
glycoprotein of Bacillus anthracis (41).
[0344] V. Methods to Search and Identify Pairs of Antigens and
Antibodies for Antigen/Antibody Microarrays
[0345] Detecting microbial antigens in serum or other body fluids
is useful in the diagnosis of an infectious disease but generally
it is difficult. It is, therefore, necessary to establish a highly
sensitive antibody-based microarray to detect microbial antigens.
In addition, such an antibody-based microarray will allow us to
monitor a microbial pathogen in the environment as well as in host
body fluids. Identification of highly specific antigens and
antibodies for each pathogen is critical for the development of a
diagnostic biochip.
[0346] We have established a simple and effective strategy to
screen for these reagents for microarray production. We printed a
panel of forty-eight carbohydrate-containing macromolecules of
distinct structural characteristics on a slide with the method
established above. In previous investigations, these antigens were
successfully applied to screen human myeloma and lymphoma proteins
for antibodies with anti-carbohydrate activities (48, 52, 53). The
carbohydrate microarray was applied to detect human serum
antibodies. A total number of twenty serum specimens were randomly
collected from normal individuals. As little as one micro-liter of
serum specimen from each individual was applied for microarray
staining. As shown in FIG. 14, twelve distinct specificities of IgM
antibodies (12/48) and thirty-five IgG anti-carbohydrate antibodies
(35/48) were identified. Carbohydrate molecules that were
positively stained include twenty polysaccharides (20/24), eleven
complex carbohydrates of cellular origin (11/19) and four
semi-synthetic glycoconjugates (4/5). For anti-carbohydrate
antibodies of IgM isotype, twelve distinct specificities were
identified. The majority (7/12) were bound to Klebsiella
polysaccharides. This is similar to our previous observation that
antibodies bound to Klebsiella polysaccharides were most frequently
found in the repertoire of human myeloma anti-carbohydrate
antibodies (48-52). The repertoire of human IgG anti-carbohydrate
antibodies is, however, broader than those of IgM isotype. There
are twenty specific for microbial polysaccharides (20/24),
including polysaccharides of microbial pathogens, E. coli K92 and
-K100; Pneumococcus type-C, -VIII, -IX, -SIV, -XIV and -27; Group B
Meningococcus, H. Influenza Type A, and different types of
Klebsiella. Human IgG antibodies specific for glycomers of complex
carbohydrate structures (11/19) and those for the carbohydrate
moieties of semi-synthetic glycoconjugates (4/5) were also
detected.
[0347] This antigen microarray was then applied to characterize
monoclonal antibodies (FIG. 14, panels III and IV) to critically
evaluate their antigenic or epitope-binding specificities and
cross-reactivities. Antigenic cross-reactivity between microbial
polysaccharide .alpha.(1,6)dextran and a preparation of Chondroitin
sulfate B polysaccharide that was derived from the intestinal
mucosa of porcine was demonstrated. This has led to the recognition
of a novel cross-reactive molecular marker of microbes and host
cell (11). In addition, these experiments demonstrated that
semi-synthetic glycoproteins are applicable for the construction of
a carbohydrate microarray. This is of critical importance since it
enables the application of rationally designed synthetic
oligosaccharides for microarray construction and allows a critical
examination of the specificity and cross-reactivity of
carbohydrate-mediated molecular recognition.
[0348] The flexible 8-chamber biochip system, as described above,
is sufficient for this type of initial screening of a panel of
molecular targets. In this format, all the available antigen
preparations of the eight Category-A pathogens can be printed at
various dilutions in a single sub-array chamber. A specific
antibody, for example, a mouse monoclonal antibody bound to the
anthrax polysaccharide, can be applied on the sub-array biochip to
react with all the antigens on the chip. It may also be applied in
a series of dilutions on different sub-array chambers. This simple
experiment allows us to critically evaluate (a) antibody binding
specificity and cross-reactivity; (b) the affinity of
antigen-antibody interaction; and (c) the quality of the printed
antigen and method for immobilizing the antigen.
[0349] If an antigen/antibody pair is qualified by the 8-chamber
biochip analysis, a further evaluation is conducted using a
large-scale biochip, which displays both microbial antigens and
those from human and other mammalian species. The highly specific
pairs of antigens and antibodies are candidates for the diagnosis
of corresponding microbial infections. Those who have
cross-reactivities with human tissue antigens may be also useful
for printing microarrays. As summarized above, identification of
such cross-reactivities may lead to a better understanding of the
biological relationship of microbes and their hosts, as well as the
pathogenesis mechanisms of an infectious disease.
[0350] VI. Methods to Search and Identify Novel Protein Targets for
Diagnosis and Vaccination
[0351] As described above, we have already established that our
microarray platform is useful for printing protein microarrays and
have demonstrated the principle and method to apply a large panel
of protein antigens to extend our scope of investigation of
microbial infections. This investigation has provided insights into
how one could incorporate the genomic information and genetic
material that has been accumulated by the genome sequencing
projects into the development of diagnostic protein biochips. We
report here the establishment of a strategy to efficiently utilize
the information from the various pathogen genome projects to assist
in the identification of novel molecular targets for diagnosis and
vaccination. The pX01 plasmid of B. anthracis will serve as a model
to illustrate our strategy.
[0352] (A) Methods to Search and Identify Candidates for the
Protein Microarray
[0353] Different categories of proteins can be used to produce the
diagnostic protein microarray. For example, (a) both membrane-bound
and secretory proteins of a pathogen, which may trigger the initial
immune response in an infection, making them ideal targets for
early diagnosis and vaccination; (b) non-surface expressed
proteins, which could be useful in identifying and characterizing
the progressive stages of an infection; and (c) the candidate genes
that are unique to a given pathogen, namely the species-specific or
stain-specific molecular targets, thereby enabling highly specific
detection on the final biochip.
[0354] Two categories of bioinformatic software, predictive and
comparative, are currently available on-line. A two-step analysis
can be used to identify candidate gene products to be included on
the microarray. First, the structure, cellular localization, and
function of newly discovered genes of the genome sequencing
projects is predicted. Then a comparative analysis to predict the
specificity and potential cross-reactivity of the candidate genes
selected by the above structure-function predictive analysis is
performed. A list of "predictive" software tools that will help in
extrapolating a protein's function are available on-line. These
include (a) TMHMM (http://www.cbs.dtu.dk/services/TMHMM-2.0/) for
the prediction of transmembrane helices in proteins; (b) NetOGlyc
2.0 (http://www.cbs.dtu.dk/services/NetOGlyc/) for the prediction
of mucin type GalNAc O-glycosylation sites in mammalian proteins;
(c) PredictProtein
(http://www.embl-heidelberg.de/predictprotein/predictprote-
in.html) for the prediction of secondary structure, solvent
accessibility, and transmembrane segments; and (d) 3D-PSSM Web
Server V 2.6.0 (http://www.bmm.icnet.uk/.sup..about.3dpssm/) for
protein fold recognition.
[0355] With the predicted location and structure of the protein now
known, either through published gene annotation done upon
completion of a genome sequence, or through predictive
bioinformatic software, it is equally important to determine
whether the gene of interest is unique to the species through a
comparative proteomic analysis. This is crucial if we hope to
design a biochip that can specifically identify if the patient is
suffering from B. anthracis. A gene that is similar in both B.
anthracis and B. subtilis would be a poor choice for the biochip,
because it would lead to inconclusive results when deployed in the
field. Therefore, careful study should be done in order to identify
if genes are highly unique to the species, which would in turn
result in a biochip with few cross-reactive ambiguous results. One
example of a potentially valuable bioinformatic tool to address
this problem is called the COGnitor program. This program takes an
amino acid sequence and extrapolates its function by comparing the
unknown protein's sequence of amino acids to amino acid sequences
from other genes of characterized function in a series of genomes
(54). Genes that are highly conserved in evolution would be
identifiable using the COGnitor program; those that are unique to a
given organism could be non-recognizable by a COGnitor search.
Proteins that are not recognized by the COGnitor program are
probably high priority candidates for diagnostic application since
the proteins they encode are most likely species-specific antigens.
For the genes that are recognized and classified by the program,
additional steps of investigation must be taken to define whether
they are suitable targets for a diagnostic use.
[0356] (B) A Model to Illustrate our Comparative Genomic Strategy
to Identify Unique Genes
[0357] The pX01 plasmid of Bacillus anthracis has been fully
sequenced, and 46 of the 142 identified genes have been
functionally characterized in the literature (16). This provides us
with a prime example of how a complete DNA sequence, with some gene
annotation, would be approached using our comprehensive
bioinformatic strategy.
[0358] First, we use predictive bioinformatic software to determine
the location and structural characteristics of the proteins from
their amino acid sequences. One example of this is in predicting if
a protein has transmembrane regions, as can be determined by TMHMM
version 2.0 (55). We have selected the protein Px01-54 as an
example. Okinaka et al. have given a description of this protein as
a "S-layer precursor/surface layer protein." This protein was run
through the transmembrane predicting software TMHMM version 2.0
(see FIG. 15).
[0359] It would seem that this protein contains both a
transmembrane region, as well as a large extracellular component.
One could hypothesize that, based on this result, the protein is on
the surface of the bacteria, and is an ideal candidate to be
printed on a microarray because it may play a role in initiating a
primary immune response. Therefore, even if Okinaka et al did not
characterize the gene, we could use this software to identify it as
an ideal surface target that could induce a primary immune
response.
[0360] With the completion of the predictive stage for function and
location of a protein, the next stage of bioinformatic analysis is
conducted, namely to determine if the gene of interest is unique to
an organism. As described above, the COGnitor program (54) can be
used to identify the species-specific genes/proteins for printing
the diagnostic protein chips. To test this idea, we have run the
amino acid sequences of all the 46 annotated genes of Px01 plasmid
of B. anthracis through the COGnitor program. We present the
results from this test in Table 2.
[0361] Of the 46 genes analyzed, 34 could be recognized and
classified according to the COGnitor program. A large number of the
COG-classified genes are of category L, namely DNA replication,
recombination and repair. Indeed, it is likely that such proteins
contain highly conserved domains across many species. At the other
extreme, twelve genes were not recognized by the COGnitor program,
and were thereby not able to be classified into the COG
classifications. These unclassifiable proteins have been listed as
"No Cog" in Table 2. This "No Cog" classification indicates that
these proteins have no similarity to other proteins currently in
the database. Since the COG database includes 74,059 classified
protein sequences from the genomes of 43 species of both
prokaryotic and eukaryotic organisms (but excludes the human
genome), a "No Cog" result suggests that the protein in question is
more likely to be species or strain-specific to the pathogen. At
the very least, one can claim that it is highly unlikely for a "No
Cog" protein to contain a known evolutionarily conserved ancient
domain that exists across many species' genomes. Being able to
select against genes that exhibit such conserved domains is a
welcome tool, for this could help to lower the probability of
selecting a protein that may yield ambiguous results on the final
biochip due to cross reactivity. Interestingly, the genes of the
most unique lethal nature, as well as a series of surface antigens,
are not recognized by the COGnitor program because they match no
other species' genes. One example of an ideal gene to select has
been presented, namely the px01-54 protein of B. anthracis. This
protein is expressed on the surface of the organism according to
Okinaka et al. (16), spans a membrane according to the TMHMM
program, and is not recognized by the COGnitor program, making it
more likely to be unique to the species.
[0362] In conclusion, combining knowledge from the published
literature and bioinformatic software on a gene product's function,
location, and `species unique` level will greatly enhance the use
of the final biochip because it will allow clear and accurate
clinical diagnosis of a patient's immune response to a wide panel
of pathogenic agents.
[0363] References
[0364] 1. Blaustein, R. O., Koehler, T. M., Collier, R. J. &
Finkelstein, A. Anthrax toxin: channel-forming activity of
protective antigen in planar phospholipid bilayers. Proc Natl Acad
Sci USA 86, 2209-2213. (1989).
[0365] 2. Thorne, C. B. Bacilluse anthracis. In Bacillus subtilis
and Other Gram-positive Bacteria, ed. A L Sonenshein, J A Hoch, R
Losick., 113-124 (1993).
[0366] 3. Escuyer, V. & Collier, R. J. Anthrax protective
antigen interacts with a specific receptor on the surface of CHO-K1
cells. Infect Immun 59, 3381-3386. (1991).
[0367] 4. Strange, R. E. & Belton, F. C. Studies on a
protective antigen produced in vitro from bacillus anthracis:
Purification and chemisty of the antigen. Bri. J. Exp. Pathol. 35,
153-165 (1953).
[0368] 5. Mourez, M. et al. Designing a polyvalent inhibitor of
anthrax toxin. Nat Biotechnol 19, 958-961. (2001).
[0369] 6. Brown, P. O. & Botstein, D. Exploring the new world
of the genome with DNA microarrays. Nature. Genetics. 21, 33-37
(1999).
[0370] 7. DeRisi, J. L., Iyer, V. R. & Brown, P. O. Exploring
the metabolic and genetic control of gene expression on a genomic
scale. Science. 278, 680-686 (1997).
[0371] 8. Ramsay, G. DNA chips: state-of-the art. Nature.
Biotechnology. 16, 40-44 (1998).
[0372] 9. MacBeath, G. & Schreiber, S. L. Printing proteins as
microarrays for high-throughput function determination. Science
299, 1760-1763 (2000).
[0373] 10. Lueking, A. et al. Protein microarrays for gene
expression and antibody screening. Analytical. Biochemistry. 270,
103-111 (1999).
[0374] 11. Wang, D., Liu, S., Trummer, B. J., Deng, C. & Wang,
A. Carbohydrate microarray leading to the recognition of
cross-reactive molecular markers of microbes and host cells. Nature
Biotechnology 20 (2002).
[0375] 12. Ekins, R. P. Multi-analyte immunoassay. J Pharm Biomed
Anal 7, 155-168 (1989).
[0376] 13. Stoll, D. et al. Protein microarray technology. Front
Biosci 7, C13-32. (2002).
[0377] 14. Volchkov, V. E. Ebola virus, complete genome. Institute
of Virology, Philipps-University (2000).
[0378] 15. Shchelkunov, S. N., Totmenin, A. V. and Sandakhchiev, L.
S. Analysis of the nucleotide sequence of 23.8 kbp from the left
terminus of the genome of variola major virus strain India-1967.
Virus Res. 40, 169-183 (1996).
[0379] 16. Okinaka, R. T. et al. Sequence and organization of pXO1,
the large Bacillus anthracis plasmid harboring the anthrax toxin
genes. J Bacteriol 181, 6509-6515. (1999).
[0380] 17. Wang, D. & Kabat, E. A. in Structure of Antigens.,
Vol. Three. (ed. M. H. V. V. Regenmortal) 247-276 (CRC Press, Boca
Raton New York London Tokyo; 1996).
[0381] 13. Wyatt, R. et al. The antigenic structure of the HIVgp120
envelope glycoprotein. Nature 393, 705-711 (1998).
[0382] 19. Robbins, J. B. & Schneerson, R.
Polysaccharide-protein conjugates: a new generation of vaccines. J.
Infct. Dis. 161, 821-832 (1990).
[0383] 20. Schneerson, R., Barrera, O., Sutton, A. & Robbins,
J. B. Preparation, characterization, and immunogenicity of
Haemophilus influenzae type b polysaccharide-protein conjugates. J
Exp Med 152,361-376. (1980).
[0384] 21. Gladstone, G. P. & Walton, E. Effect of iron on the
bactericidal proteins from rabbit polymorphonuclear leukocytes.
Nature 227, 849-851. (1970).
[0385] 22. Ezzell, J. W., Jr., Abshire, T. G., Little, S. F.,
Lidgerding, B. C. & Brown, C. Identification of Bacillus
anthracis by using monoclonal antibody to cell wall
galactose-N-acetylglucosamine polysaccharide. J Clin Microbiol 28,
223-231. (1990).
[0386] 23. Finne, J., Leinonen, M. & Makela, P. H. Antigenic
similarities between brain components and bacteria causing
meningitis. Implications for vaccine development and pathogenesis.
Lancet. 2, 355-357 (1983).
[0387] 24. Finne, J. & Makela, P. H. Cleavage of the
polysialosyl units of brain glycoproteins by a bacteriophage
endosialidase. Involvement of a long oligosaccharide segment in
molecular interactions of polysialic acid. Journal. of. Biological.
Chemistry. 260, 1265-1270 (1985).
[0388] 25. Mandrell, R. E. et al. Lipooligosaccharides (LOS) of
some Haemophilus species mimic human glycosphingolipids, and some
LOS are sialylated. Infection. &. Immunity. 60, 1322-1328
(1992).
[0389] 26. Mandrell, R. E. Further antigenic similarities of
Neisseria gonorrhoeae lipooligosaccharides and human
glycosphingolipids. Infection. &. Immunity. 60, 3017-3020
(1992).
[0390] 27. Feizi, T. & Loveless, R. W. Carbohydrate recognition
by Mycoplasma pneumoniae and pathologic consequences. Am. J.
Respir. Crit. Care. Med. 154, S133-136. (1996).
[0391] 28. Burrel, C. J. Serological markers of hepatitis B. Clin
Gastroenterol 9, 47-63. (1980).
[0392] 29. Boxall, E. H., Peterson, S., Diment, J., Graham, S. E.
& Shirley, J. A. A novel assay for hepatitis B e markers. J Med
Virol 52, 280-285. (1997).
[0393] 30. Basten, A. & Howard, J. G. in Contemporary Topics in
Immunobiology, Vol. 2. (ed. A. J. S. Davies) 265 (Plenum, New York;
1973).
[0394] 31. Humphrey, J. H., Parrott, D. M. V. & East, J.
Studies on globulin and antibody, production in mice thymectomised
at birth. Immunology 7, 419-439 (1964).
[0395] 32. Weissman, I. L., Gutman, G. A., Friedberg, S. H. &
Jerabek, L. Lymphoid tissue architecture. III. Germinal centers, T
cells, and thymus-dependent vs thymus-independent antigens. Adv.
Exp. Med. Biol. 66, 229-237 (1976).
[0396] 33. Beagley, K. W., Black, C. A., Dunkley, M. L. &
McGhee, J. R. in Cytokine Regulation of Humoral Immunity. (ed. C.
M. Snapper) 391-408 (John Wiley & Sons Ltd, Chichester, New
York, Brisbane,Toronto, Singapore; 1996).
[0397] 34. McGhee, J. R., Mestecky, J., Elson, C. O. & Kiyono,
H. Regulation of IgA synthesis and immune response by T cells and
interleukins. [Review]. Journal of Clinical Immunology 9, 175-199
(1989).
[0398] 35. Mestecky, J. The common mucosal immune system and
current strategies for induction of immune responses in external
secretions. [Review]. Journal of Clinical Immunology 7, 265-276
(1987).
[0399] 36. Mongini, P. K. A., Stein, K. E. & Paul, W. E. T-cell
regulation of IgG subclass antibody production in response to
T-independent antigens. J. Exp. Med. 153, 1-12 (1981).
[0400] 37. Kagnoff, M. F. & Murray, P. D. T dependent induction
of an IgA and IgM anti-polysaccharide response. Advances in
Experimental Medicine & Biology, 155-167 (1987).
[0401] 38. Ivars, F., Nyberg, G., Holmberg, D. & Coutinho, A.
Immune response to bacterial dextrans. II. T cell control of
antibody isotypes. Journal of Experimental Medicine 158, 1498-1510
(1983).
[0402] 39. Ehrhardt, R. O. et al. Differential activation
requirements of isotype-switched B cells. Eur. J. Immunol. 26,
1926-1934 (1996).
[0403] 40. Tonkinson, J. L. & Stillman, B. A. Nitrocellulose: a
tried and true polymer finds utility as a post-genomic substrate.
Front Biosci 7, C1-C12. (2002).
[0404] 41. Cole, H. B., Ezzell, J. W., Jr., Keller, K. F. &
Doyle, R. J. Differentiation of Bacillus anthracis and other
Bacillus species by lectins. J Clin Microbiol 19, 48-53.
(1984).
[0405] 42. Cohen, S. et al. Attenuated nontoxinogenic and
nonencapsulated recombinant Bacillus anthracis spore vaccines
protect against anthrax. Infect Immun 68, 4549-4558. (2000).
[0406] 43. Smith, H. & Zwartouw, H. T. The polysaccharide from
Bacillus anthracis grown in vivo. Biochem J 63, 447-452 (1956).
[0407] 44. Kabat, E. A., Bezer, A. E., Baer, H. & Knaub, V.
Immunochemical studies on blood groups. VII. Chemical changes
associated with destruction of blood group activity and enhancement
of the type XIV cross-reactivity by partial hydrolysis of hog and
human blood group A, B, and O, substances. J. Exp. Med. 88, 43-57
(1948).
[0408] 45. Kabat, E. A., Howe, C. & Schiffman, G.
Immunnochemical studies on blood groups. J. Am. Chem. Soc. 80, 6656
(1958).
[0409] 46. Kabat, E. A. Immunochemical studies on blood groups. J.
Immunol. 82, 358-372 (1958).
[0410] 47. Mesnage, S. et al. Bacterial SLH domain proteins are
non-covalently anchored to the cell surface via a conserved
mechanism involving wall polysaccharide pyruvylation. Embo J 19,
4473-4484. (2000).
[0411] 48. Kabat, E. A. et al. Human monoclonal macroglobulins with
specificity for Klebsiella K polysaccharides that contain
3,4-pyruvylated-D-galactose and 4,6-pyruvylated-D-galactose. J.
Exp. Med. 152, 979-995 (1980).
[0412] 49. Kabat, E. A. et al. The epitope associated with the
binding of the capsular polysaccharide of the group B meningococcus
and of Escherichia coli K1 to a human monoclonal macroglobulin, IgM
November J. Exp. Med. 168, 699-711 (1988).
[0413] 50. Kabat, E. A., Liao, J., Sherman, W. H. & Osserman,
E. F. Immunochemical characterization of the specificities of two
human monoclonal IgM's reacting with chondroitin sulfates.
Carbohydr. Res. 130, 289-297 (1984).
[0414] 51. Kabat, E. A., Liao, J., Shyong, J. & Osserman, E. F.
A monoclonal IgM lambda macroglobulin with specificity for
lacto-N-tetraose in a patient with bronchogenic carcinoma. J.
Immunol. 128, 540-544 (1982).
[0415] 52. Kabat, E. A. et al. A human monoclonal macroglobulin
with specificity for a(2-8)-linked poly-N-acetyl neuraminic acid,
the capsular polysaccharide of group B meningococci and Escherichia
coli K1, which crossreacts with polynucleotides and with denatured
DNA. J. Exp. Med. 164, 642-654 (1986).
[0416] 53. Nickerson, K. G., Liao, J. & Kabat, E. A. A
monoclonal IgM kappa from a blood group B individual with
specificity for alpha-galactosyl epitopes on partially hydrolyzed
blood group B substance. Carbohydr. Res. 243, 345-357 (1993).
[0417] 54. Tatusov, R. L., Koonin, E. V. & Lipman, D. J. A
genomic perspective on protein families. Science 278, 631-637.
(1997).
[0418] 55. Krogh, A., Larsson, B., von Heijne, G. & Sonnhammer,
E. L. Predicting transmembrane protein topology with a hidden
Markov model: application to complete genomes. J Mol Biol 305,
567-580. (2001).
[0419] Fourth Series of Experiments
[0420] HydroGel.TM. slides are produced by Packard BioScience for
printing protein microarrays. We investigated whether Hydrogel can
serve as an alternative substrate for producing carbohydrate
microarrays and for producing HIV diagnostic protein
microarrays.
[0421] I. Materials and Methods
[0422] A. Procedures for printing and Staining of the
Hydrogel-Based Biochips
[0423] (1) Method for Printing on Hydrogel-Coated Slides
[0424] The method for printing carbohydrates and proteins on
Hydrogel-coated slides is essentially identical to that for
printing on nitrocellulose slides. The only difference is that the
pre-treatment of Hydrogel follows the method described by the
manufacture as follows: (a) place HydroGel slides in 40.degree. C.
incubator for 20 minutes; and (b) remove prior to printing and
allow slides to cool to room temperature for 5 minutes.
[0425] 2) Method for Post-Printing Treatment
[0426] The procedures for post-printing treatment were modified
from the manufacture's instruction as follows: (a) incubate arrays
overnight in a humidified 30.degree. C. chamber; (b) rinse briefly
(for 20 seconds) in PBST (PBS+0.05% Tween 20), followed by 3, 30
minute washes with PBST. Rinse for 20 seconds in PBS (pH 7.4); and
(c) dry slides by centrifugation at 1600 RPM for 5 minutes in a
table-top centrifuge.
[0427] (3) Method for Staining
[0428] The methods for staining Hydrogel biochips are identical to
those for staining nitrocellulose biochips.
[0429] II. Results and Discussion
[0430] Applicants have shown the following surprising results:
[0431] 1) Printing proteins and carbohydrate-containing
macromolecules on hydrogel can be performed using Cartesian's
PIXSYS 5500A Microarryer (CHIPMAKER 4), using a contact arraying
procedure.
[0432] 2) Protein preparations, including antibodies, BSA, Avidin
and HTV-1 gp120, RT and gag proteins, can be quantitatively
immobilized on the hydrogel, providing fine spots of about 150
microns in diameter and thus making it possible to produce a high
density protein microarray.
[0433] 3) Carbohydrate preparations, including polysaccharides,
glycosaminoglycans, glycoproteins, semi-synthetic glycoconjugates
and glycolipids, can be quantitatively immobilized on the hydrogel,
providing fine spots of about 200 microns in diameter and thus
making it possible to produce a high density carbohydrate
microarray.
[0434] 4) The immobilized protein and carbohydrate antigens
analyzed have their immunological properties well-preserved at the
time the slides were stained using specific antibodies.
[0435] 5) Use of the Hydrogel-based protein and carbohydrate
microarrays is advantageous due to their low fluorescent
background. A disadvantage of the Hydrogel substrate is its
absorption of relative lower amounts of materials on the biochip.
For the same antigen preparations, the fluorescent signals detected
on Hydrogel-slides are much lower than those detected on the
nitrocellulose slides.
[0436] 6) When the ratios of fluorescent intensity over background
of given microspots are calculated, the results of
nitrocellulose-chips and those of Hydrogel-chips are closely
correlated in most cases (see Table 4 for a comparison among
antigens of distinct structural characteristics).
[0437] 7) The nitrocellulose-based biochip favors detection of IgG
antibodies but not IgM antibodies in solution. The former is an
immunoglobulin (Ig) monomer; and the latter is an Ig pentamer. Such
bias is not seen in the Hydrogel-based biochips. This finding is of
significance since it provides information to improve the
nitrocellulose surface for producing a diagnostic biochip.
[0438] Fifth Series of Experiments
[0439] I. Introduction
[0440] An infectious agent may expose and release multiple
antigenic substances to a host, eliciting specific antibody
responses. Many antigen-antibody binding assays are currently in
use for clinical diagnosis of infectious and non-infectious
diseases. These include the classical direct immunoassays, such as,
immunodiffusion, immunoelectrophoresis, agglutination and
immunoprecipitation, and recently developed methods, including
immunofluorescence, radioimmunoassay (RIA), enzyme-immunoassay
(EIA) and western blot. These approaches take advantage of the
specificity of antigen-antibody interaction but are designed to
operate on a one-by-one basis. Their successful diagnostic
application for infectious diseases is, therefore, largely
dependent on whether a clinician makes the correct decision
regarding which diagnostic test should be run.
[0441] Establishment of technologies to enable the simultaneous
detection and characterization of a wide range of microbial
infections represents a current challenge to the worldwide effort
of biodefense. High-priority infectious agents that pose current
risks to our national security include multiple microbial
pathogens, such as Bacillus anthracis (anthrax), Clostridium
botulinum (botulism), Yersinia pestis (plague), Variola viruse
(smallpox), Francisella tularensis (tularemia) and viral
hemorrhagic fevers (Lassa virus, Ebola virus, Marburg virus and LCM
virus). An assay system that allows detection and characterization
of these pathogens in a single assay is currently unavailable.
Rapid progress of the genome sequencing projects has brought new
hopes as well as challenges to biodefense efforts. For example, a
great deal of genomic information is available for numerous
pathogens.sup.1,2,3. This leads the way to developing a generation
of high throughput technologies to combat emerging infectious
diseases. These include the nucleic acid-based microarrays.sup.4,5
or DNA chips.sup.6, and the protein-based microarrays.sup.7,8.
Application of these technologies in the field of infectious
diseases is still in its early stages.sup.9,10. Novel ideas are
badly needed for the establishment of high throughput technologies
to monitor potential bioterrorism attacks as well as naturally
occurring emerging infectious diseases.
[0442] Our laboratory has been focusing on the development of a
biochip platform that is suitable for the production of both
carbohydrate and protein-based microarrays. Our carbohydrate
microarray technology is described herein. Such a technology is
designed to display a large repertoire of pathogen-specific
antigens of distinct structures in order to monitor a broad range
of microbial infections. Since it is applicable for human and many
other animal species, this system is particularly useful for
monitoring the emergence of infectious diseases. In many infectious
diseases, an intermediate reservoir plays important roles in the
life cycle of the pathogen. In other cases, both human and animal
species are accessible to a given pathogen. A number of category A
pathogens (Bacillus anthracis, Clostridium botulinum, Yersinia
pestis, and viral hemorrhagic fevers) and other emerging infectious
diseases (Lyme disease and West Nile Virus infection) have such
characteristics. Thus, the antigen/antibody-based biochips will
play a significant role in the surveillance of emerging infectious
diseases (FIG. 17).
[0443] Both carbohydrate antigens and protein antigens are
important for the production of a diagnostic biochip. As shown in
FIG. 17, the presence of a polysaccharide in Bacillus anthracis has
been recognized for some time.sup.11,12 13. It is specific for B.
anthracis, a useful molecular target for the identification of the
pathogen. This structure is, however, masked before spore
germination, casting doubt on its potential for vaccination use. A
protein antigen of B. anthracis, the protective antigen (PA) is a
well-recognized vaccine target. PA is an integrated component of
the lethal toxin of Bacillus anthracis. It binds to a specific
cellular receptor and forms cell bound complexes with edema factor
(EF) and lethal factor (LF).sup.14-16. Understanding of this
process led to the development of polyvalent inhibitor-based
therapeutic strategies to protect a host from lethal attack by the
toxin.sup.17.
[0444] Given our current knowledge of protein chemistry, on-chip
protein structural alterations, including denaturation or partial
denaturation and various degrees of conformational modification,
are unavoidable in the course of a long time period in storage
(months to years). However, the such alterations are acceptable
when a protein chip is made for an immunological application
because a naturally occurring infection or vaccination elicits
specific antibodies directed to both the native and the
denatured/partially denatured antigenic determinants.
[0445] In summary, detecting a microbial antigen and/or its
specific antibodies in bodily fluids are important in diagnosis of
infectious diseases and in surveillance of emerging infectious
diseases. This goal is achieved by an antigen/antibody-based
microarray of high sensitivity, large capacity, and extensive
antigenic diversity.
[0446] II. Procedures for Production of Long-Lasting Protein
Microarrays
[0447] This section is directed to a central technical difficulty
in the technology of proteomic microarrays--the structural and
conformational instability of protein on-chip. Our procedures
include a) a high throughput technology to screen for the antigen
preparations that are applicable for our proposed microarray
platform; b) the technology for the production and long-lasting
preservation of protein microarrays; and c) Green Chip Technology
to facilitate protein chip development.
[0448] (A) High Throughput Screening to Identify Stable Antigens
for the Production Protein Chips
[0449] In our preliminary investigation, we found that the on-chip
stability of protein antigens varies among molecules and/or
preparations. Molecular mechanisms underlying these differences are
quite complex. To meet the urgent need of biodefense, we take a
straightforward but efficient approach to facilitate the
development of diagnostic biochips. Specifically, we have
established a microarray-based high throughput screening technology
to identify suitable microbial antigens for biochip production.
[0450] Our method of protein chip fabrication is a non-covalent
protein immobilization on a nitrocellulose-coated micro-glass
slide. It uses a nitrocellulose-surface to immobilize an antigen,
similar to the Western Blot assay, but has no treatment to denature
the protein before its application. A protein microspot is,
therefore, printed on the chip without any chemical or physical
treatment to interfere with its native structure.
[0451] We have designed a single step assay in which protein
antigens are printed on the nitrocellulose-coated slides and then
stained by antibodies to "see" whether specific antigenic
determinants are displayed on chip and whether they are accessible
to antibodies in solution. The monoclonal and polyclonal antibodies
elicited by a microbial antigen or by a pathogen, as well as serum
specimens of an infected individual, are all useful reagents for
these analyses. A positively stained microspot indicates the
presence of a target molecule or antigenic determinant.
[0452] We have established two methods, a) an eight-chamber
sub-array system (FIG. 10) and b) a full-scale biochip scanning
(FIG. 11), to facilitate the screening process.
[0453] 1. An Eight-Chamber Sub-Array System
[0454] Each microglass slide contains eight well-separated
sub-arrays of identical contents. Its microarray capacity is 600
microspots per subarray. A single slide is, therefore, designed to
enable eight microarray analyses. Antibodies can be applied on
separate subarrays in a serial dilution as in an ELISA assay.
[0455] a) 8-Chamber Sub-Arrays (See FIG. 10):
[0456] Each microglass slide contains eight-well separated
sub-arrays of identical contents. There are 600 microspots per
subarray, with spot sizes of approximately 200 microns and at
300-micron intervals, center-to-center. A single slide is,
therefore, designed to enable eight detections.
[0457] b) Contains:
[0458] i) A 600-spot subarray is composed of pathogen-specific
protein antigens in question as well as a few selected carbohydrate
antigens to serve as control spots. These selected carbohydrate
antigens have been characterized on the chip. Their specific
antibodies are seen frequently in body fluids of normal
individuals;
[0459] ii) Repeats and dilutions: Each antigen will be printed at
0.5-1.0 mg/ml and also at a one to ten dilution of the initial
concentration for the second concentration. Each preparation at a
given concentration will be repeated three times; and
[0460] iii) Antibody isotype standard curves: Human or murine
antibodies of IgG, IgA and IgM isotype of known concentrations will
serve as standard curves for antibody detection and
normalization.
[0461] For each antigen preparation, approximately twenty-five
micrograms of purified protein are sufficient for producing this
set of testing chips. We designed the following protocol in order
to enable a rapid and economic screening: a) Negative elimination;
b) Specificity and cross-reactivity; and c) Conventional
immunoassay. Preparations of mixed antibodies are made in such a
way that a) the total protein concentrations of the mixed
antibodies for each subarray staining is identical across
subarrays; b) for each antibody preparation, there are series
dilutions (four points); c) antibodies of same species are grouped
to avoid frequently occurring cross-neutralization of antibody
reactivity among different species and to allow detection of
multiple Ig isotypes on the same spots.
[0462] a) Negative elimination: A preparation of mixed monoclonal
or polyclonal antibodies is applied to stain a subarray. Antibodies
derived from the same species can be mixed together to stain a
sub-array and a tagged secondary antibody can reveal their specific
binding on antigen spots. To take the advantage of the above
8-chamber design, these antibodies are applied in serial dilutions
to permit an estimation of their detection sensitivity and
specificities;
[0463] b) Specificity and cross-antigenic reactivity: A subarray of
the antigen chip is stained using distinctly tagged monoclonal or
affinity purified antibodies to identify specific antigen spots.
This staining allows one to detect specificity as well as potential
cross-antigenic reactivity among printed antigens. Such analysis
has been detailed previously; and
[0464] c) Conventional immunoassays to verify microarray results:
ELISA, Dot blot, and/or Western Blot is performed. If an antigen is
positively detected by an antibody on ELISA but negative or poorly
detectable by protein microarray, this may suggest a problem with
its on-chip immobilization. The protein may be conjugated to a
carrier molecule (such as BSA) to improve its immobilization. If it
is ELISA negative but Western positive to a corresponding antibody,
the protein may be solubilized in a denaturing solution before
printing on the chip.
[0465] 2. A Full-Scale Biochip Scanning
[0466] We have produced a microarray biochip composed of about
4,000 microspots of antigens. In addition to carbohydrate antigens,
we printed a large panel of protein antigens. In this design, each
antigen or antibody preparation of a given dilution was printed
with four repeats in a vertical line of micro-spots on the biochip.
This allowed us to visually observe and statistically analyze the
reproducibility of microarray printing and staining. The
significance and sensitivity of antibody detection for each antigen
at a given antigen concentration can also be statistically
calculated. Some preparations, for example the gp120 glycoprotein
of HIV-1 as highlighted in a square in FIG. 11, were printed from
left to right in a series dilution of one to five, beginning at
0.5-1.0 mg/ml and with four dilutions thereafter. These biochips
were stained with the uninfected or HIV-1-infected human serum
specimens with the methods described above. Pictures of the
ScanArray visualization of the multi-color fluorescent staining of
biochips are shown in FIG. 11.
[0467] (B) Technologies for the Production and Preservation of
Long-Lasting Protein Microarrays
[0468] Given our understanding of protein chemistry, on-chip
protein structural alteration, including denaturing or partial
denaturation and various degrees of conformational modification, is
unavoidable over a long period of time (months to years). Such
alteration is generally un-acceptable when a protein chip is made
to study the biological function of a protein, such as
protein-protein interactions, small molecule transportation,
enzymatic reactivities, and cell signaling activity, etc. To
maintain these functional properties of proteins, substantial
preservation of the native protein structure or conformation is
generally necessary. For an immunological application, this
situation is, however, significantly different since an antibody
may recognize either the native or the denatured/partially
denatured antigenic determinants. Antibodies of distinct
epitope-binding specificities are frequently elicited in the course
of a microbial infection. Technically, the ELISA-based immunoassays
are preferential to the detection of antibodies that recognize the
native antigenic determinants; the Western blot assays generally
directed to the protein structures that are denatured/partial
denatured and immobilized on a nitrocellulose membrane or other
surfaces with hydrophobic characteristics. The "Western type" of
antigenic structures also has large potential in protein chip
application since many antigens on Western blot/nitrocellulose
strips are stable and suitable for long-term preservation.
[0469] Differing from the DNA-based microarray assay, which
requires DNA and/or RNA denaturation to allow the A/T or C/G
hybridization-taking place, the protein microarray assay requires
preservation of protein structures (a general belief in the field).
Protein on-chip stability has been considered a critical issue for
the development of long-lasting protein microarrays. To address
this question of protein stability, we analyzed in a model system,
the HIV-1 protein chip (see FIG. 18).
[0470] FIG. 18 shows the results from protein microarrays that were
printed, air-dried, and stored at room temperature. In one set,
slides were stained at day 2 and another set in 11 months after
printing. Serum specimens applied for the day 2 staining are
identical to those for the 11 month-old slides. This allows a
comparative analysis of protein antigenic reactivities on the chips
that were stored for different time range, either for two days or
for eleven months. Each bar represents the mean of four detections
(Ratio of intensity over background). Data were calculated and
statistically analyzed as described.sup.9.
[0471] This experiment showed that some protein preparations, such
as SF2 gp120-glycosylated and GST-gag p24 of HIV-1 preserved their
antigenic reactivities quite well after a prolonged storage in
air-dried condition at room temperature. Other preparations,
including SF2gp120 non-glycosylated, LAV gp120 glycosylated,
GST-gag p55 and GST-gag p55/mutant P222A, lost a large proportion
of their antigenic reactivities on chip after long-term storage.
These observations indicate an existing possibility that production
of a long-lasting protein chip using our microarray platform is
most likely feasible.
[0472] 1. Long-Lasting Preservation of Global Patterns of Antibody
Reactivities On-Chips
[0473] To test further this possibility, we characterized a large
collection of antigen preparations. Specifically, we stained three
sets of biochips that were made identically in the same printing
experiment using same antibody preparations at three time points,
day 2, eight months, and eleven months after they were made and
simply stored at room temperature in air. A clustering
analysis-based software was developed by our lab to characterize
the antigenic reactivities on-chip in a global manner. Results
illustrated in FIG. 19 reveal that although some proteins showed
reduced antigenic reactivities on-chip after a long-term storage in
air at room temperature, the global patterns of antigenic
reactivities remain the same or similar among three different
experiments across the three time points: day 2, 8 months and 11
months.
[0474] In three replicate experiments, biochips were stained at day
2 (Experiment A), 8 months (Experiment B), or 11 months (Experiment
C), using the same specimens, either serum of an uninfected
individual or an HIV-1 infected patient. Results were presented as
the Log ratio of the intensity over background for each antigen
microarray at a given concentration. A hierarchical clustering
method is applied on the data under the platform of Matlab. The
correlation between color bar and the log ratio of intensity over
background is indicated in FIG. 19 (right corner).
[0475] 2. Methods of Post-Printing Handling to Improve Protein Chip
Stability
[0476] We have established a few methods for the post-printing
treatment of protein chips to improve the stability of our protein
chip. Our methods are relatively simple and generally applicable to
many protein antigens, including those that have distinct
structural characteristics. This is, in fact, a challenge to the
proteomic microarray technology. Most, if not all protein antigens,
express two general types of antigenic determinants, being either
labile or stable on chip. To improve the stability of diagnostic
protein chips, two different types of approaches may be considered:
methods to stabilize the native structures of protein on chip and
methods to remove the labile epitopes and/or to enrich the stable
epitopes. We established, therefore, the following approaches to
establish a long-lasting antigenic protein microarray:
[0477] a) Lyophilizing Protein On-Chip
[0478] Many protein preparations can be lyophilized and stably
stored at low temperature for a long period of time. We treat our
printed protein chips by a step of lyophilizing. We then seal them
in a plastic bag and store them at lower temperatures, including
4.degree. C., -20.degree. C., and -80.degree. C. These chips are
analyzed as described above after a period of storage (two days,
six months and two years). Results are compared with those of
experiment 1 whereby the printed chips are placed at room
temperature in the air-dried condition. A technical concern is
whether the process of lyophilizing sucks off a proportion of
printed proteins. If a significant signal reduction is seen, a
period air-drying the freshly made chip can be employed to avoid
the signal reduction.
[0479] b) Dessication of Protein Microarrays:
[0480] Place a protein chip in a desiccator for a period of five to
ten days in the presence of a desiccating reagent without a vacuum.
Then stain the chip with antibodies to probe corresponding
antigenic determinants. We predict that some but not all antigenic
structures/protein epitopes are sensitive to dehydration. ELISA and
Western blot assay are conducted to further characterize the two
classes of antigenic structures, the dehydration sensitive and the
dehydration resistant antigenic determinants.
[0481] c) Hydration of Protein Microarrays:
[0482] Prepare three sets of protein chips by covering them with a
thin layer of "sticky" aqueous solution, i.e., 40% glycerol in
1.times. PBS. One set (six chips) is stored at 4.degree. C., one
set at -20.degree. C. and the last set at -80.degree. C. After a
period of storage (two days, six months, and two years), two
chips/sixteen sub-arrays are stained and the data analyzed as
described above.
[0483] In Western blot diagnosis of HIV-1 and many other infectious
diseases, serum antibodies of infected individuals are found in
many cases to be specific to denatured or partially denatured
protein antigens. We, however, for the first time, take advantage
of this host immunity to develop a novel protein chip assay.
Differing from Western blot, the protein chips can substantially
extend the spectrum of pathogens that can be detected in a single
assay. In addition, a chip assay is rapid, high throughput, and
economic. It requires only a few microliters of serum specimen.
Theoretically, similar structural alteration may also occur in vivo
in the cellular events of antigen processing/trafficking during a
natural infection, generating antigenic structures that differ from
those expressed by the native proteins. Such structures are still
"foreign" to a host and capable of induction of specific antibody
responses. The potential of this category of antigenic structures
for diagnostic use, especially in association with the high
throughput protein microarray technology, is yet to be
explored.
[0484] In summary, protein microarrays can be treated in three
different ways, including lyophilizing, desiccation, and hydration,
and stored for a various periods of time, ranging from two days to
two years.
[0485] (C) Green Chip Technology--the Second Generation of Protein
Chips
[0486] As described above, we have established technologies to
explore the available antigens and antibodies to rapidly establish
protein antigen-based microarrays to meet the urgent needs of
biodefense. The potential of these approaches is, however, limited
by the availability of pathogen-specific antigens and antibodies.
Here, we apply the state-of-art technologies of protein engineering
to facilitate the development of protein microarrays. Specifically,
we a) introduce the technology of green fluorescent protein (GFP)
in our next generation of protein microarrays, the Green Chip
Technology; b) establish a molecular engineering-based strategy to
produce recombinant proteins that fluoresce on the chip and are
capable of displaying stable antigenic determinants on the surface
of a microarray. This method can be applied to many molecular
targets whose nucleotide sequences and/or structural information
are available. This method is, therefore, designed to extend the
repertoire of pathogen signatures for the production of diagnostic
microarrays. We use the protective antigen (PA) of B. anthracis as
a model to demonstrate this technology.
[0487] Efficacy of protein immobilization on a
nitrocellulose-coated microslide depends on the physicochemical
property of protein molecules. Since only noncovalent molecular
interactions are involved, a common influencing factor is the
molecular weight (MW) of a target molecule. In some circumstances,
protein antigens or pathogen-specific antigenic determinants are
too small to be immobilized on the chip without altering their
antigenic reactivities. We, therefore, fuse a target molecule to a
carrier protein molecule to improve its surface adhering property.
In this case, we select a fluorescent carrier molecule, the green
fluorescent protein (GFP). GFP was originally isolated from the
biolyminescent jellyfish Aequorea Victoria.sup.44. The protein, as
well as its variants, have been successfully expressed in various
other organisms and have been used as a detection system. The two
variants that are readily available as purified proteins are EGFP
(enhanced green fluorescent protein) and GFPuv (UV-optimized GFP
variant). Their excitation maxima are 488 and 395 nm, respectively,
while their emission maxima are 507 and 509 nm, respectively.
[0488] We tested purified GFPuv on our protein microarray chip. It
generates bright fluorescent signal upon excitation at fluorescence
channel (488 nm) but not at Cy3 (552 nm) and Cy5 (633 nm) chanels
of a microarray scanner (GSI SA5000) (FIG. 20 and table 5). This
makes it practical to use the FITC channel to monitor protein
stability, printing quality, and data normalization and, while
using Cy3, Cy5 and other GFP negative channels to detect
antigen-antibody interaction on the same spots. In addition, the
GFP signal is retained on the chip after it is air-dried and
extensively washed (FIG. 20). Introduction of GFP to the protein
chip `up-grades" our protein chip technology. It provides not only
an effective and convenient means to monitor the quality of protein
microarrays but also an internal control for each microspot. The
latter is greatly important for data processing and data mining and
therefore will generally improve the quality of protein chips.
[0489] 1. The PA Antigen of B. anthracis as a Model for the Green
Chip Technoloy
[0490] PA is the central component of the three-part lethal toxin
of Bacillus anthracis. It binds to a specific cellular receptor and
forms toxic, cell bound complexes with edema factor (EF) and lethal
factor (LF).sup.14-16. It was named PA since it acted protective
when an animal was injected with the protein and then challenged by
the lethal pathogen B. anthracis. PA itself is a good antigen for
eliciting antibody responses in a number of mammalian species,
making it an important vaccine against anthrax.sup.45. The crystal
structure of monomeric PA at 2.1 .ANG. resolution and the
water-soluble heptamer at 4.5 .ANG. resolution have been resolved
(FIG. 21).sup.46. PA is, therefore, an ideal molecule model for
studying the relationship between protein structure and its
antigenic reactivity.
[0491] Our investigation is directed at the recognition of the
antigenic determinants of PA that are stable on a chip, useful for
vaccine evaluation, and highly specific for pathogen recognition.
We focus on the relationship between structure and the on-chip
antigenic reactivity of the molecule. Specifically, we ask a)
whether the structures that are buried in its native configuration
are useful for protein chips, and b) whether the structures that
are displayed on the surface in the native form are stable and long
lasting on chip. The structures hidden in a native protein may
become exposed in vivo by cellular or enzymatic processing or in
vitro owing to its on-chip structural alteration. Our working
hypothesis is that such hidden structures are previously
unrecognized pathogen signatures and are suitable for the
production of long-lasting diagnostic protein chips. In PA, an
important "hidden structure" is the surface of PA.sub.63, which is
masked by the N-terminal 20K fragment of PA (PA.sub.20). Release of
this fragment from PA leads to heptamer formation by PA.sub.63.
[0492] Our overall experimental design uses antibodies of distinct
epitope binding specificities to recognize the antigenic
determinants of PA, to monitor their possible alteration on chip
during storage, and thereby identify the stable antigenic
structures for long-lasting protein chips. Experimentally, we a)
produce a panel of recombinant proteins of PA and its derivatives;
b) produce the domain-specific PA protein microarrays using these
recombinant PAs; c) immunize mice using PA and its conjugates to
elicit antibody responses and monitor the time course of the
response by PA protein chips; and d) generate monoclonal antibodies
to ensure the availability of antibodies of various binding
specificities, including those that bind PA specifically by Western
Blot and those specific for the antigen on ELISA.
[0493] 2. GFP-Fusion Proteins of PA, its Derivatives, and Their
Fluorescent protein Microarrays
[0494] a) Construction of expression vectors, protein induction and
purification: Construct a panel of twelve GFP-PA fusion proteins as
outlined in FIG. 22. Briefly, six sets of PCR primers have been
designed according the published nucleotide sequences of PA (on the
PX01 plasmid) and the domain structure of the protein. Use a cDNA
clone containing the full-length PA as a template for PCR
amplification to produce six PCR fragments, including a full-length
PA, an N-terminal truncated PA (PA.sub.63), and domain 1 to 4.
[0495] These PCR fragments are subcloned into an E. coli expression
vector, pGFPuv cDNA Vector (BD/Clontech), to express corresponding
GFP fusion proteins. In vector design, the expressed fusion protein
will be accumulated intracellularly as GFP alone. In addition, a
His-tag can be cloned either at the N-terminal or the C-terminal of
a recombinant protein, allowing protein purification using Ni-NTA
column/resin under either denatured or native condition as
described by the manufacture (QIAGEN Inc.,USA). This will enable
the binding of protein on Ni-NTA ELISA plate and Western blot using
anti-His antibodies. Additional methods of purification include: i)
the native condition of purification; ii) denaturing conditions;
and iii) the GFPuv protein is resistant to the 8M Urea, which
serves as one of the alternative methods for GFP-fusion protein
purification.
[0496] b) Printing and validating the fluorescing PA microarrays:
Methods for microarray printing, preservation, scanning, and data
processing are the same as described above. Given our current
knowledge, these GFP-fusion proteins are as bright as GFP alone. If
some clone loses its fluorecent property, anti-GFP antibody can be
used to detect them on the chip. Proteins may be solublized using a
solution containing 8 M urea. We observed previously that proteins
solubilized in this way can be immobilized on a nitrocellulose
membrane or on an ELISA plate. Printing is on the 8-chanber
subarray system as described above. Protein chips are stored under
different conditions as described above and analyzed two days, six
months, and one year after printing. Antibody staining is detailed
below using a panel of monoclonal and polyclonal antibodies.
[0497] c) Animal immunization and production of monoclonal
antibodies: A panel of polyclonal and monoclonal anti-PA antibodies
is currently available. These reagents are tested by ELISA and by
Western blot to identify different categories of specificities: a)
Recognize native epitope: ELISA strongly positive, but Western
negative or weakly positive to PA; b) Recognize denatured/partially
denatured epitope: Western strongly positive, but ELISA negative or
weakly positive; and c) Recognize the stable epitopes: Both methods
strongly positive. These are relative ways of classification but
have been proven useful for predicting an antigenic structure.
[0498] 3. "Green Chip Technology" to Reveal the Hidden Antigenic
Determinants
[0499] As described above, we believe that the on-chip protein
structural alteration, including denaturation or partial
denaturation and various degrees of conformational modification, is
unavoidable in the course of a long period of time (months to
years). Such alteration is, however, not necessarily a limiting
factor to the diagnosis of infectious disease and other
immunological disorders on a protein chip. We reason that
antibodies elicited by an antigenic stimulation, either by a
microbial infection or by a vaccination, have the potential to
recognize multiple antigenic structures, including either the
native/conformational sensitive antigenic determinants or the
denatured or partially denatured/conformationally insensitive
antigenic determinants. For instance, the ELISA-based immunoassays
preferentially detect antibodies that recognize the native
antigenic determinants; the Western blot assays are directed to the
protein structures that are denatured/partially denatured and
immobilized on a nitrocellulose membrane or other surfaces with
hydrophobic characteristics. "Western-like" antigenic structures
have high potential in protein chip application since they are
generally stable and suitable for long-term preservation. Given the
complexity of protein structure and the question regarding their
antigenic reactivity, we use the protective antigen (PA) of B.
anthracis as a model and example for other important microbial
antigens.
[0500] Experimentally, we a) take advantage of our green chip
technology to monitor the structural stability and quality of
printed protein chips by capturing its GFP signal on the chip over
periods of time; b) use a large panel of monoclonal and polyclonal
antibodies specific for distinct antigenic determinants of PA,
including specificities to the native structures as well as
reactivities directed to the denatured/partially denatured
structures. This type of on-chip epitope scanning is high
throughput. We perform it at defined points in time as described
above.
[0501] III. A Procedure for Production of Epitope-Specific
Carbohydrate Microarrays
[0502] This section is directed to the high throughput production
of the oligosaccharide-based, epitope-specific carbohydrate
microarrays. This invention makes it possible to produce a
glycochip whereby a single type of glycoconjugate, the
oligosaccharide-polyacrylomide conjugate, is utilized to display a
large repertoire of oligosaccharides of distinct structures.
[0503] (A) .alpha.Gal Anti-.alpha.Gal System as a Model
[0504] We have chosen the .alpha.Gal anti-.alpha.Gal system as a
model to establish the technology of the oligosaccharide-based,
epitope-specific carbohydrate microarrays. Since aGal is a key
antigenic determinant that is responsible for the hyperacute
rejection in animal to human xenotransplantation, the microarrays
produced by this investigation are valuable for monitoring such
anti-xenoantigenic responses.
[0505] We printed both .alpha.Gal-protein conjugates (data not
shown) and .alpha.Gal-polyacrylamide conjugates on nitrocellulose
coated glass slides and then stained the microarray with
ugal-specific antibodies. Since we have already established that
glycoproteins can be stably immobilized on the nitrocellulose
slide, the use of .alpha.Gal-protein conjugates in this study
serves as a control for neoglycopolymers.
[0506] 1. Microarray Experiment-1: Lectin Staining
[0507] To validate whether the .alpha.Gal-sugar epitopes are
displayed by the PAGE-conjugates, we spotted a panel of
.alpha.-Gal-PAGE conjugates of defined structures on a
nitrocellulose-coated glass slide.
[0508] We spotted PAGE-based neoglycoconjugates at an initial
concentration of 0.6 mg/ml and then diluted the conjugates in
serial dilutions of 1:4. We stained the microarray chips using a
lectin that is specific for the sugar chain. The Banderiraea
Simplicifolla BS-1 lectin (Sigma) is used in FIG. 23A. The lectin
was coupled with a biotin molecule. The microspots-captured lectin
was thus revealed by a streptoavidin-Cy5 conjugate. Glycoconjugates
utilized in this study include: (i) DW450 phenyl
.alpha.-Galactopyranosyl-(1-3)
b-D-galactopyranosyl-(1-4)-1-thio-b-D-glucopyranoside; (ii)
DW451.alpha.-Galactosyl-2; (iii) DW452 .alpha.-Galactosyl-3; (iv)
DW453 .alpha.-Galactosyl-4; (v) DW454 polylactose (1/2.5); (vi)
DW455 .alpha.-Gal polymer A; (vii) DW456 .alpha.-Gal polymer B;
(viii) DW457 .alpha.-Gal polymer C; (ix) DW458 .alpha.-Gal polymer
D; (x) DW459 .alpha.-Gal polymer E; and (xi) DW460 .alpha.-Gal
polymer F.
[0509] Results shown in FIG. 23A demonstrate that the
.alpha.Gal-sugar epitopes are displayed by the PAGE-conjugates and
are specifically detected by the lectin Banderiraea Simplicifolla
BS-1.
[0510] 2. Experiment-2: Human Antibody Characterization of the
.alpha.Gal-Glycochips
[0511] To investigate further whether the PAGE-conjugates of
oligosaccharides are applicable for our platform of microarray
production, we stained the microarrays with a preparation of
affinity purified human anti-.alpha.-Gal antibodies at a 1:5
dilution (FIG. 23B upper) and at 1:50 (FIG. 23B bottom). The
captured human IgG antibodies were then stained with an
anti-Hu-IgG-Biotin conjugate, and subsequently stained using Av-Cy5
at the 1:250 dilution. Data for three repeats of the experiment
were statistically analyzed.
[0512] Glycoconjugates utilized in this study include: (i) DW450
phenyl .alpha.-Galactopyranosyl-(1-3)
b-D-galactopyranosyl-(1-4)-1-thio-b-D-gluc- opyranoside; (ii)
DW451.alpha.-Galactosyl-2; (iii) DW452 .alpha.-Galactosyl-3; (iv)
DW453 .alpha.-Galactosyl-4; (v) DW454 polylactose (1/2.5); (vi)
DW455 .alpha.-Gal polymer A; (vii) DW456 .alpha.-Gal polymer B;
(viii) DW457 .alpha.-Gal polymer C; (ix) DW458 .alpha.-Gal polymer
D; (x) DW459 .alpha.-Gal polymer E; and (xi) DW460 .alpha.-Gal
polymer F.
[0513] Results shown in FIG. 23B illustrate that the
.alpha.Gal-sugar epitopes are displayed by the PAGE-conjugates and
are specifically detected by a preparation of affinity purified
human anti-.alpha.Gal antibodies. This experiment provides direct
information regarding the use of this sugar chip in detecting human
anti-.alpha.Gal antibodies. Therefore, this sugar chip itself is a
means for monitoring the hyperacute rejection in animal-to-human
xenotransplantation that is mainly mediated by human
anti-.alpha.Gal antibodies. The potential of this epitope-specific
sugar chip is, however, far beyond this specific application.
[0514] References
[0515] 1. Volchkov, V. E. Ebola virus, complete genome. Institute
of Virology, Philipps-University (2000).
[0516] 2. Shchelkunov, S. N., Totmenin, A. V. and Sandakhchiev, L.
S. Analysis of the nucleotide sequence of 23.8 kbp from the left
terminus of the genome of variola major virus strain India-1967.
Virus Res. 40, 169-183 (1996).
[0517] 3. Okinaka, R. T. et al. Sequence and organization of pX01,
the large Bacillus anthracis plasmid harboring the anthrax toxin
genes. J Bacteriol 181, 6509-6515. (1999).
[0518] 4. Brown, P. O. & Botstein, D. Exploring the new world
of the genome with DNA microarrays. Nature. Genetics. 21, 33-37
(1999).
[0519] 5. DeRisi, J. L., Iyer, V. R. & Brown, P. O. Exploring
the metabolic and genetic control of gene expression on a genomic
scale. Science. 278, 680-686 (1997).
[0520] 6. Ramsay, G. DNA chips: state-of-the art. Nature.
Biotechnology. 16, 40-44 (1998).
[0521] 7. MacBeath, G. & Schreiber, S. L. Printing proteins as
microarrays for high-throughput function determination. Science
299, 1760-1763 (2000).
[0522] 8. Lueking, A. et al. Protein microarrays for gene
expression and antibody screening. Analytical. Biochemistry. 270,
103-111 (1999).
[0523] 9. Wang, D., Liu, S., Trummer, B. J., Deng, C. & Wang,
A. Carbohydrate microarray leading to the recognition of
cross-reactive molecular markers of microbes and host cells. Nature
Biotechnology 20 (2002).
[0524] 10. Grifantini, R. et al. Previously unrecognized vaccine
candidates against group B meningococcus identified by DNA
microarrays. Nat Diotechnol 20, 914-921. (2002).
[0525] 11. Gladstone, G. P. & Walton, E. Effect of iron on the
bactericidal proteins from rabbit polymorphonuclear leukocytes.
Nature 227, 849-851. (1970)
[0526] 12. Strange, R. E. & Belton, F. C. Studies on a
protective antigen produced in vitro from bacillus anthracis:
Purification and chemisty of the antigen. Bri. J. Exp. Pathol. 35,
153-165 (1953).
[0527] 13. Ezzell, J. W., Jr., Abshire, T. G., Little, S. F.,
Lidgerding, B. C. & Brown, C. Identification of Bacillus
anthracis by using monoclonal antibody to cell wall
galactose-N-acetylglucosamine polysaccharide. J Clin Microbiol 28,
223-231. (1990).
[0528] 14. Blaustein, R. O., Koehler, T. M., Collier, R. J. &
Finkelstein, A. Anthrax toxin: channel-forming activity of
protective antigen in planar phospholipid bilayers. Proc Natl Acad
Sci USA 86, 2209-2213. (1989).
[0529] 15. Thorne, C. B. Bacilluse anthracis. In Bacillus subtilis
and Other Gram-positive Bacteria, ed. A L Sonenshein, J A Hoch, R
Losick., 113-124 (1993).
[0530] 16. Escuyer, V. & Collier, R. J. Anthrax protective
antigen interacts with a specific receptor on the surface of CHO-K1
cells. Infect Immun 59, 3381-3386. (1991).
[0531] 17. Mourez, M. et al. Designing a polyvalent inhibitor of
anthrax toxin. Nat Biotechnol 19, 958-961. (2001).
[0532] 18. Deng, C., Wang, A., Wang, D. & Zhang, P. in Talks on
DIMACS Workshop on Analysis of Gene Expression Data.
DIMACSPiscataway, N.J.; 2001).
[0533] 19. Deng, C., Zhang, P., Wang, A., Trummer, B. & Wang,
D. in Proceedings of IEEE International Joint Conference on Neural
NetworksHawaii, U.S.A; 2002).
[0534] 20. Ramdas, L. et al. Sources of nonlinearity in cDNA
microarray expression measurements. Genome Biol 2 (2001).
[0535] 21. Kim, S. et al. General nonlinear framework for the
analysis of gene interaction via multivariate expression arrays. J
Biomed Opt 5, 411-424. (2000).
[0536] 22. Tseng, G. C., Oh, M. K., Rohlin, L., Liao, J. C. &
Wong, W. H. Issues in cDNA microarray analysis: quality filtering,
channel normalization; models of variations and assessment of gene
effects. Nucleic Acids Res 29, 2549-2557. (2001).
[0537] 23. Jain, A. K., Duin, R. P. W. & Mao, J. Statistical
Pattern Recognition: A Review. IEEE TRANSACTIONS ON PATTERN
ANALYSIS AND MACHINE INTELLIGENCE 22, 4-37 (2000).
[0538] 24. Nadler, M. & Smith, E. P. Pattern recognition
engineering. (John wiley & Sons Inc., New York; 1993).
[0539] 25. Bow, S.-T. Pattern recognition and image preprocessing.
(Marcel Dekker, Inc, New York; 1992).
[0540] 26. Alon, U. et al. Broad patterns of gene expression
revealed by clustering analysis of tumor and normal colon tissues
probed by oligonucleotide arrays. Proc Natl Acad Sci USA 96,
6745-6750. (1999).
[0541] 27. Ben-Dor, A. Clustering gene expression patterns. Journal
of Computational Biology 6, 281-297 (1999).
[0542] 28. Heyer, L. J., Kruglyak, S. & Yooseph, S. Exploring
expression data: identification and analysis of coexpressed genes.
Genome Res 9, 1106-1115. (1999).
[0543] 29. Carr D B, Somogyi R & G S, M. Templates for Looking
at Gene Expression Clustering. Statistical Computing and Graphics
Newsletter 8, 20-29 (1997).
[0544] 30. Michaels, G. S. et al. Cluster analysis and data
visualization of large-scale gene expression data. Pac Symp
Biocomput, 42-53. (1998).
[0545] 31. Eisen, M. B., Spellman, P. T., Brown, P. O. &
Botstein, D. Cluster analysis and display of genome-wide expression
patterns. Proc Natl Acad Sci USA 95, 14863-14868. (1998).
[0546] 32. Butte, A. J. & Kohane, I. S. Mutual information
relevance networks: functional genomic clustering using pairwise
entropy measurements. Pac Symp Biocomput, 418-429. (2000).
[0547] 33. DeRisi, J. et al. Use of a cDNA microarray to analyse
gene expression patterns in human cancer. Nat Genet 14, 457-460.
(1996).
[0548] 34. Schena, M. Quantitative monitoring of gene expression
patterns with a complementary DNA microarray. [see comments.].,
(1995).
[0549] 35. Schena, M. et al. Parallel human genome analysis:
microarray-based expression monitoring of 1000 genes. Proc Natl
Acad Sci USA 93, 10614-10619. (1996).
[0550] 36. Vapnik, V. N. Statistical learning theory. (John Wiley
& Sons, New York; 1998).
[0551] 37. Vapnik, V. N. An overview of statistical learning
theory. IEEE TRANSACTIONS ON NEURAL NETWORKS 10, 988-999
(1999).
[0552] 38. Burges, C. J. C. A tutorial on support vector machines
for pattern recognition. Data Mining and Knowledge Discovery 2,
121-167 (1998).
[0553] 39. Su, A. et al. Molecular classification of human
carcinomas by use of gene expression signatures. Cancer Res. 61,
7388-7393 (2001).
[0554] 40. Brown, M. P. S. et al. Knowledge-based analysis of
microarray gene expression data by using support vector machines.
PNAS 97, 262-267 (2000).
[0555] 41. Furey, T. S. et al. Support vector machine
classification and validation of cancer tissue samples using
microarray expression data. Bioinformatics 16, 906-914. (2000).
[0556] 42. Ramaswamy, S. et al. Multiclass cancer diagnosis using
tumor gene expression signatures. Proc Natl Acad Sci USA 98,
15149-15154. (2001).
[0557] 43. Fried, R., Gather, U. & Imhoff, M. Online pattern
recognition in intensive care medicine. Proc AMIA Symp, 184-188.
(2001).
[0558] 44. Chalfie, M. Green fluorescent protein. Photochem
Photobiol 62, 651-656. (1995).
[0559] 45. Reuveny, S. et al. Search for correlates of protective
immunity conferred by anthrax vaccine. Infect Immun 69, 2888-2893.
(2001).
[0560] 46. Petosa, C., Collier, R. J., Klimpel, K. R., Leppla, S.
H. & Liddington, R. C. Crystal structure of the anthrax toxin
protective antigen. Nature 385, 833-838. (1997).
[0561] 47. Peterson, L. Factor analysis of cluster-specific gene
expression levels from cDNA microarrays. Comput Methods Programs
Biomed 69, 179. (2002).
1TABLE 1 Structural properties of dextrans and Inulin preparations
Plain dextrans FITC conjugates N279 LD7 B1299S Dex-20K Dex.-70K
Dex-2000K Inulin Structural characteristics Molecular weight (kDa)
.about.10,000 42 .about.20,000 19.6 71.2 2,000 3.2 Sugar residue
Glucose Glucose Glucose Glucose Glucose Glucose Fruclose
Conformational characteristics Linear Linear Heavily Linear Linear
Linear Linear chain chain branched chain chain chain chain dominant
only dominant dominant dominant dominant (?) Molar ratio
(FITC:sugar) 0 0 0 0.01 0.005 0.008 0.007 Proportions of linkages
(%).sup.I .alpha.(1,6) Terminal nonreducing end group 5 0 31 5 5 5
0 .alpha.(1,6) Backbone 90 100 32 90 90 90 0
.alpha.(1,3).alpha.(1,6- ) Backbone 0 0 1 0 0 0 0
.alpha.(1,2).alpha.(1,6) Backbone 0 0 5 0 0 0 0 .alpha.(1,3)
Branches 5 0 1 5 5 5 0 .alpha.(1,2) Branches 0 0 30 0 0 0 0
.beta.(2,1) Linkage 0 0 0 0 0 0 100 (?) .sup.IValues are estimated
from methylation and periodate-oxidalion analysis (see ref. 18 for
a summary).
[0562]
2TABLE 2 Microarray detection and characterization of human and
murine anti-carbohydrate antibodies.sup.a Antigen micro-spots I.
Human IgM II. Human IgG Antigen name Class ID Location Mean SD
Int./BK Mean SD Int./BK Klebsiella type 7 1 1 A1 19293 1786 1.12
32140 3367 3.86 Klebsiella type K11 1 2 A2 19560 3349 1.13 15262
7630 1.52 Klebsiella type K13 1 3 A3 39103 4354 2.17 25997 719 3.20
Klebsiella type K21 1 4 A4 22847 2131 1.27 29255 890 3.63 Dudmans
Rhizobium TA1 1 5 A5 31625 2768 1.77 16198 693 2.06 Chondrotin SO4
"B" 2 6 A6 17009 633 0.96 10830 411 1.40 Pneumococcus type C 1 7 C1
17014 1661 0.98 25187 3499 3.02 Pneumococcus type VIII 1 8 C2 17194
1407 0.98 12075 3754 1.47 Pneumococcus type XIV 1 9 C3 17012 1262
0.96 12292 4286 1.52 Cow 21 (Blood group B) 3 10 C4 19336 2180 1.08
9280 839 1.15 Bacto-agar 20 .degree. C. Ext 1 11 C5 19792 2512 1.09
9682 1276 1.20 Arabino Galactan (Larch) 1 12 C6 16216 453 0.90 8599
474 1.08 IM3-BSA.sup.d 4 13 E1 19743 1898 1.11 14322 1750 1.70
IM3-KLH.sup.d 4 14 E2 18077 1185 1.01 10541 930 1.27 Le.sup.a
(N-110% 2x) 3 15 E3 17046 728 0.95 8960 605 1.08 Beach P1 (Blood
group B) 3 16 E4 17419 525 0.97 10022 423 1.21 Tii II (Blood group
B & Le.sup.a)) 3 17 E5 17946 464 0.99 9400 503 1.14 Ogd 3 18 E6
20682 2620 1.13 8819 459 1.08 ASOR.sup.c 3 19 G1 16555 449 0.93
8933 320 1.06 LNT-BSA.sup.c 4 20 G2 19053 1266 1.05 9010 287 1.08
Phosphomannan 1 21 G3 17571 785 0.97 16124 923 1.93 Meningococcus
group B 1 22 G4 16747 620 0.93 8839 403 1.06 H Infl Type A 1 23 G5
17804 656 0.98 11007 208 1.33 E. coli K92 1 24 G6 17353 770 0.95
8785 328 1.07 Klebsiella type A3 1 25 I1 27018 9910 1.40 13001 6947
1.94 Klebsiella type K12 1 26 I2 32322 16450 1.69 11539 5029 1.72
Klebsiella type K14 1 27 I3 23360 1283 1.22 54557 2045 7.80
Klebsiella type K33 1 28 I4 54607 2574 2.60 22890 1259 3.37
Chondrotin SO4 "A" 2 29 I5 18093 1252 0.99 7646 730 1.16 Chondrotin
SO4 "C" 2 30 I6 17699 983 0.97 7547 607 1.14 Pneumococcus type SIV
1 31 K1 19002 662 1.00 11485 819 1.68 Pneumococcus type IX 1 32 K2
18932 1303 1.00 10600 3827 1.56 Pneumococcus type 27 1 33 K3 23308
6222 1.24 13455 7158 1.97 Cow 26 (Blood group B) 2 34 K4 19184 1743
1.03 8020 885 1.19 Helo Pomaba Galactan 1 35 K5 17885 722 0.97 7298
550 1.11 Helo Nemoralis Galactan 1 36 K6 18135 715 0.98 7412 416
1.10 IM6-BSA.sup.d 4 37 M1 18897 184 1.01 9991 784 1.41
IM6-KLH.sup.d 4 38 M2 17761 289 0.96 7657 544 1.13 Le.sup.a (N-1
IO.sub.4NaOH) 3 39 M3 18430 926 1.00 7857 626 1.11 Cyst 9 (Blood
group A) 3 40 M4 18182 552 0.99 9081 733 1.30 Dextran N-150-N (60K)
1 41 M5 17527 717 0.96 7908 671 1.17 Hog (Blood Group H) 3 42 M6
17791 1015 0.98 7706 506 1.13 AGOR.sup.d 3 43 O1 18218 305 0.98
7698 485 1.09 Inulin 1 44 O2 17655 361 0.95 7636 326 1.08 Levan
(B-512E) 1 45 O3 18003 318 0.97 11807 2116 1.68 Meningococcus group
Y 1 46 O4 17276 841 0.95 7749 433 1.09 E. coli K1 1 47 O5 17518
1219 0.97 7442 508 1.11 E. coli K100 1 48 O6 17852 807 0.99 15823
2152 2.33 Background (n = 200) 18267 844 7522 727 Total number of
positives: 12 35 Antigen micro-spots III. Anti-Dex 4_3F1 IV.
Anti-Dex 16.4.12E Antigen name Class ID Location Mean SD Int./BK
Mean SD Int./BK Klebsiella type 7 1 1 A1 9074 215 1.01 11432 324
1.02 Klebsiella type K11 1 2 A2 9584 837 1.06 11432 262 1.03
Klebsiella type K13 1 3 A3 23003 3573 2.56 12256 648 1.10
Klebsiella type K21 1 4 A4 8817 203 0.98 12487 367 1.12 Dudmans
Rhizobium TA1 1 5 A5 8438 448 0.93 10901 226 0.98 Chondrotin SO4
"B" 2 6 A6 59264 822 6.38 18063 935 1.62 Pneumococcus type C 1 7 C1
8813 373 0.98 11256 271 1.00 Pneumococcus type VIII 1 8 C2 8862 362
0.99 11450 512 1.02 Pneumococcus type XIV 1 9 C3 8879 330 0.99
11359 375 1.02 Cow 21 (Blood group B) 3 10 C4 9020 325 1.00 11309
450 1.02 Bacto-agar 20 .degree. C. Ext 1 11 C5 9106 251 1.02 11269
527 1.02 Arabino Galactan (Larch) 1 12 C6 8564 433 0.97 10909 357
0.99 IM3-BSA.sup.d 4 13 E1 10490 258 1.17 65535 0 5.34
IM3-KLH.sup.d 4 14 E2 8629 390 0.98 62128 5069 5.42 Le.sup.a
(N-110% 2x) 3 15 E3 8742 287 1.00 11026 501 0.98 Beach P1 (Blood
group B) 3 16 E4 8638 307 0.96 10926 561 0.97 Tii II (Blood group B
& Le.sup.a)) 3 17 E5 8556 202 0.95 11006 457 0.98 Ogd 3 18 E6
8912 427 1.00 11148 705 0.99 ASOR.sup.c 3 19 G1 8627 547 0.96 10923
566 1.00 LNT-BSA.sup.c 4 20 G2 8509 497 0.96 11000 316 0.99
Phosphomannan 1 21 G3 8585 483 0.96 11224 394 1.01 Meningococcus
group B 1 22 G4 8633 285 0.96 11115 623 0.99 H Infl Type A 1 23 G5
8637 376 0.95 11205 499 1.01 E. coli K92 1 24 G6 8714 396 0.97
11248 190 1.01 Klebsiella type A3 1 25 I1 10278 1120 1.12 11496 136
1.01 Klebsiella type K12 1 26 I2 9303 293 1.01 11504 226 1.01
Klebsiella type K14 1 27 I3 9687 343 1.04 11752 405 1.03 Klebsiella
type K33 1 28 I4 16135 3620 1.70 11285 424 1.00 Chondrotin SO4 "A"
2 29 I5 9443 451 1.01 11322 215 1.00 Chondrotin SO4 "C" 2 30 I6
11936 3057 1.35 11810 396 1.05 Pneumococcus type SIV 1 31 K1 9305
351 1.03 11388 94 1.00 Pneumococcus type IX 1 32 K2 9345 697 1.02
11407 227 1.01 Pneumococcus type 27 1 33 K3 9104 914 1.00 11394 252
1.00 Cow 26 (Blood group B) 2 34 K4 9296 916 1.01 11309 205 1.00
Helo Pomaba Galactan 1 35 K5 9185 917 0.99 11369 268 1.01 Helo
Nemoralis Galactan 1 36 K6 9179 773 0.98 11234 185 1.00
IM6-BSA.sup.d 4 37 M1 10610 423 1.13 62509 2863 5.43 IM6-KLH.sup.d
4 38 M2 9303 250 1.00 13534 460 1.19 Le.sup.a (N-1 IO.sub.4NaOH) 3
39 M3 9223 216 0.98 11293 240 1.00 Cyst 9 (Blood group A) 3 40 M4
9129 333 0.98 11571 689 1.04 Dextran N-150-N (60K) 1 41 M5 57385
1630 6.02 12653 197 1.13 Hog (Blood Group H) 3 42 M6 9604 288 1.01
11526 145 1.04 AGOR.sup.d 3 43 O1 9228 384 0.99 11327 360 1.00
Inulin 1 44 O2 9274 391 0.99 11264 511 1.00 Levan (B-512E) 1 45 O3
9493 190 1.00 11423 521 1.01 Meningococcus group Y 1 46 O4 9158 80
0.97 11454 687 1.01 E. coli K1 1 47 O5 9193 222 0.97 11236 364 1.00
E. coli K100 1 48 O6 9157 163 0.96 11178 270 1.00 Background (n =
200) 9161 356 11252 365 Total number of positives: 4 4 .sup.aDots
of four microarrays were statistically analyzed and the positive
results were highlighted in Bold Italics. For human serum antibody
staining, a positive score is given if the mean fluorescent
intensity value of a micro-spot is significally higher than the
mean background of the identically stained microarray with the same
fluorescent color. For the staining using monoclonal antibodies, a
positive score is given if the mean fluorescent # intensity value
of a micro-spot is at least 1.5 folds higher than the mean
background. .sup.bCarbohydrate antigens were classified and
indicated in table as the follows 1 for Polysaccharide, 2 for
Glycosaminoglycan: 3 for Glycoprotein; and 4 for Semi-synthetic
Glycoconjugate. .sup.cInt./BK: Ratio of mean fluorescence intensity
to mean background. .sup.dAGOR: Agalacto-ontaomucoid; IM isomaltose
oligossochande; # KLH, keyhole lampet hemocyanin; LNT:
lacto-N-tetraose; OG Ogunsheye 10% 2X (Blood group I activity)
[0563]
3TABLE 3 COGnitor Analysis of Annotated Genes on the PX01 Plasmid
of B. anthraces. Cog Information Calculated Independantly and Chart
from Okinaka et al Journal of Bachanology Oct. 1999 p. 6609-6516.
Added to the Chart CRF Start Stop Strand Size(m) Description Cog
Chase Cog Category 137 173709 173694 1 61 Hypothetical protein
similar to host factor protein & (65 sdl, ymaH B ou R
Uncharacterized ACR, host factor & protein 139 174581 174871 2
96 Hypothetical protein (138 aaj, uvgU; 8 substris (Z99121); 5996
ss postl O Disulide horid formulatoin protein Dsco B 138 174200
174493 2 97 Small DNA binding, pagR-ulue; B anthracis (AF031382);
53198 ss positi K Predicted transcriptional regulators 109 131838
132228 2 99 Similar to small DNA binding proteins, peoR; pasemid
pXO1, B, snevaci K Predicted transcriptional regulators 121 153808
153637 2 110 Adonine phosphorbosyl transterape, apt, plasmid pXO1;
B enthracis (s F Aderunalguarone phosphorlbosy transferases and
related PRPS 120 183848 184238 1 130 Thuricated sartiposass For
161627, B anthracis ]U30712]; 63/137 as pc L Putative transposses
116 149232 149664 2 150 Unknown function, plasmid pXO1, B,
anthracis (L13641) No Cog No Cog 87 101962 102444 2 160 Unknown
function, positive theoredorten (137 as), y011; B. subties (29911
OC Theol-desumds isomerase and thioredoxins 120 153084 152836 1 190
Putative transposase (401 gal; S. pyogenes (AF064540); 108/482 as p
L Transposase 115 142410 142991 2 193 Roenivase (191 ss) Tn1548
blue Entersooccus faschm (AO8237); 125 L S le-specific
recontinsions, DNA invertasePin homologs 114 138229 120843 2 204
Hypothetical protein in the protective antigen domain, yps, plasmid
pXC No Cog No Cog 127 162232 162876 2 214 Putative transposase for
181627, plasmid pXO1; B. anthraoe (U30712) L Putative transposase
141 175663 178307 2 214 Thermonuclease precursor (TNASEnmicrococcal
nuclease (231 am), 5 L Micrococcal nuclease (Thermonuclease)
hompiloge 85 99838 100319 1 227 Hypothetical protein (244 aa), ydL
in bitr spouC intergenic region; B su R Predicted metal- dependent
membrane protease 130 165317 166030 1 237 Hypothetical protein (251
aa) yrpE. B. subolis (U83875), 213/251 aa pc No Cog No Cog 98
116307 117131 1 274 Pulative tranposase for IS1627, plasmid pXO1; 8
anthrace Steare L L Pulstive transposase 91 109000 109642 1 280
Hypothetical protein (193 aa) ywoA m ng8.spotO intergens region 8 I
Membrane associated phospholipid phosphatase 94 112518 113403 1 296
UOP glucose-pyrophosphoragse (202 as) gtaB, B subtils (Q058520 ; M
UOP-glucose pyrophosphoryase 16 25124 26071 1 315
Integrated/recombruse protein (311 as) U therrosulotrophcum AEO L
Integrase 103 123018 123819 1 317 Protoative integrassrecombriase
(296 aa) nok, B substil (P46352) 93 L Integrase 38 48012 49689 2
325 Transocessa (478 aa) IS23IE. 6 shurengensis (002403), 282/299
221 L Predicted transposase 112 138540 139523 2 327 Spare
germination response gorKC plasmid pXO1.B anthraces (AF101 No Cog
No Cog 114 141002 142061 2 369 Spare germination response gerX8
plamsid pXO1.B anthraces (AF10 No Cog No Cog 132 167948 169033 1
361 Positive megrase (358 ss) Streptococcue musens (AF065141)
206/348 S ncharacterized BCR 136 172265 17380 1 364 Response
regulator aspartste phosphatase C (383 am), rapC, O subtilus
(060433) 126/238 as positive (32%) and 52/117 as positive (44%) 93
111374 112474 1 366 Hyaturonase synthase (419 22), gsA,
Strepcococcus pyogenes IL2118 M Gycosytransperas, probably involved
in oam wal tiogenasis 86 100377 101822 2 381 Hypothetical protein
(402 aa), m&Q B suntise (D84432). 198/376 aa pc R
Uncharacterized membrane protein 54 67634 66819 1 193 S-layer
precursor surface layer protein (814 aa); Q armraos (P49081k No Cog
No Cog 45 57659 58985 2 435 Cell division protein (372 aa), AsZ P
nortioshi (AP0000001), 103/213a D Cell division GT Pase 96 113430
114761 1 413 Putative NDP-sugar dehydrogenase (440 aa), yagf; 9
subtles (Z99121 M Predicted UDP glucose 6-dehydrogenase 23 31621
33006 2 461 Potential coding region (609 aa), Clostridium ourscle
(X96508k 248/390 L Restrontype raverge transcriptase 81 95878 97363
1 461 Rue-and transposion-related protein (225 aa), rocA B. subtise
(AF007E M Membrane proteins nutralized to metalitidopeptidas 110
150042 151469 1 475 trase acting positive regulator AuA, plasmid
pXO1, B. anthraca (L1384 No Cog No Cog 59 74068 75501 1 477
Secratory protein kinase (474 aa), kin; C arricola (U77780);
166/310 as N Type IV secretory pethway, Vin611 components, and
related AT 35 42103 43539 1 476 Transpoasse (476 aa) IS231E; B
Shuffgensis (OO2403), 456/476 aa I L Predicted transposase 38 43806
46262 2 464 Transoosase (476 aa) 45231E B. thunngerus (OO2403);
347/468 aa I L Predicted transposase 113 139480 140058 2 402 Spore
germination response gerXA plasmid pXO1, B anthracis (AF10 No Cog
No Cog 7 6554 0190 2 546 Reverse transcriptase (574 aa), IS620, E.
coli (A611549); 306/343 aa p L Retron-type reventer transcriptase
00 105712 106730 1 652 Hypothetical protein (380 as) PFBO765er
Pasmodum taloparum (AEC L ATFase involved in DNA repair 110 133381
135455 2 764 Antrax loction protecous antigen, pegA, formerly peg,
plasmid pXO1 B. No Cog No Cog 122 154224 156628 2 800
Calmedutin-sanitive adenyale cyclaseiedema factor cya; plasmid pXC
D Chromosome segregation ATPadues 107 127442 129871 1 809 Antrax
laxin lethal factor ur, plasmid pXO1; B anthracs (U129061 and No
Cog No Cog 142 178738 179398 2 867 Toposorhefase 1 lop 1 plasmid
pXO1 B anthrace (AI97227) L Topoisomerase IA 110 143196 146110 1
972 Transposase for IN21 (968 as), trpA E. coli (P13094), 519/915
as pos No Cog No Cog 79 91443 95111 1 1,222 Hypothetical
hydrophcolic protein (567 aa) Bacllus Entile (U54516); 12 N
Methyl-accepting chemclaos protein 13 15040 19002 2 1,320
Erythrocyte Invasionthoptry protein (2,401 aa) P. yosti (U36927),
620/ No Cog No Cog
[0564]
4TABLE 4 Nitrocellulose and hydrogel as substrates for carbohydrate
and protein microarrays Ratio of fluorescent intensity over
background of microspots FAST Slides Hydrogel Human IgG specific
for Mean SD Sum Mean SD Sum 1_Polysaccharides 5.35 8.28 6383.02
5.17 10.16 6163.61 (2-150) n = 149 2_Glycosaminoglycan 1.74 1.08
181.25 1.81 0.63 188.30 (GAG) 152-164) n = 13 3_Glycoprotein 2.27
3.36 1836.22 2.79 4.88 2254.38 (166-266) n = 101 4_Semi-synthetic
1.57 1.02 313.67 2.01 1.63 402.20 glycoconjugates (248-292) n = 45
5_Glycolipides 1.45 0.34 139.52 1.68 0.19 161.18 (249-305) n = 58
6_HIV-1 proteins 4.32 8.40 6051.71 2.61 6.19 3659.95 (307-481) n =
175 7_Other proteins 3.57 6.95 2428.71 3.44 6.47 2340.21 (483-567)
n = 85 8_Ig 4.27 8.11 2186.80 6.94 38.21 3554.90 (569-632) n = 64
BK_Empty space 1.24 0.11 951.36 1.61 0.44 1237.51 selected
(634-719) n = 96 BK_Saline sports (test 1.44 0.54 553.80 1.86 1.88
716.07 for cross- contamination) (731-778) n = 48 Total (n = 834)
3.42 5.58 2633.25 3.37 9.89 2588.30
[0565]
5TABLE 5 GFPuv generates bright fluorescent signal upon excitation
at 488 nm Excitation Emission Channel (nm) (nm) Positive Negative
dexsy 335 526 .DELTA. FAM 488 508 .star. Alexa488 488 522 .star.
FluorX 488 522 .star. JOE 488 532 .star. FITC 488 522 .star.
Alexe532 543 570 .DELTA. TAMRA 543 570 .DELTA. Cy3 543 570 .DELTA.
Alexa546 543 570 .DELTA. TMR 543 570 .DELTA. Alexa568 543 578
.DELTA. ROX 594 614 .DELTA. TexasRed 594 614 .DELTA. Alexa594 594
614 .DELTA. Cy5 633 670 .DELTA. APC 650 660 .DELTA.
[0566]
Sequence CWU 1
1
12 1 19 DNA Artificial Sequence Primer 1 atggttcttt agctttctg 19 2
22 DNA Artificial Sequence Primer 2 ctatcctatt ccattaagat cc 22 3
18 DNA Artificial Sequence Primer 3 gcgaagtaca agtgctgg 18 4 20 DNA
Artificial Sequence Primer 4 ctcctgttaa cgtgtaagtt 20 5 20 DNA
Artificial Sequence Primer 5 tcaggcagaa gttaaacagg 20 6 18 DNA
Artificial Sequence Primer 6 gggaacaccg tcgaatag 18 7 18 DNA
Artificial Sequence Primer 7 tggcagctta tccgattg 18 8 22 DNA
Artificial Sequence Primer 8 gcgtttaagt tctttgttga cg 22 9 21 DNA
Artificial Sequence Primer 9 caagaaacaa ctgcacgtat c 21 10 22 DNA
Artificial Sequence Primer 10 caattcctcc gagtatctct tc 22 11 19 DNA
Artificial Sequence Primer 11 tatcaagaat cagttagcg 19 12 21 DNA
Artificial Sequence Primer 12 ccgatactct atcctattcc a 21
* * * * *
References